Signaling status : the role of music in the preference for loud luxury

SIGNALING STATUS: THE

ROLE OF MUSIC IN THE

PREFERENCE FOR LOUD

LUXURY

Master Thesis

Final version

MSc. in Business Administration - Marketing

Student: Arleny Viola

Student number: 11416890

Thesis Supervisor: Tina Dudenhöffer

Date: 20th of June 2017

1

S

TATEMENT OF

O

RIGINALITY

This document is written by Arleny Viola who declares to take full responsibility for the contents of this document.

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

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

2

A

BSTRACT

Some say that the clothes people wear is a form of self-expression. Others say that the music people listen to expresses what they feel. Is there a connection between choice of clothing and choice of music then? This thesis takes a closer look at the role that music plays in the preference for luxury apparel. As the luxury market is driven by status - an individual’s prestige position in relation to others - the focus in this thesis has been on low-income consumers as these consumers tend to have a greater need for status. Furthermore, this greater need for status has a strong correlation with liking loud luxury – luxury items with prominent logos – and with emulating behavior. Thus, apparel loud luxury is the segment that is considered as the greater visibility of apparel is thought to have a greater impact on this emulating behavior.

A distinction was made between highbrow – more sophisticated music – and lowbrow – easy to listen to music – music genres. The consumers preferring lowbrow music genres were expected to have a higher preference for loud apparel luxury, especially if they had a greater need for status. However, if they had a greater appreciation for diverse music styles (musical omnivorousness), this effect would be weakened. On the other hand, the consumers liking highbrow music genres were expected to have a relatively lower preference for loud apparel luxury unless they had a greater need for status. An experiment through an online survey was conducted after which 240 respondents provided relevant data. The respondents were asked to indicate their music preferences and were afterwards exposed to different quiet and loud luxury items they could choose from.

Results showed that there was indeed a significant association between the preference for highbrow or lowbrow music genres and the preference for loud apparel luxury. Primarily, those liking highbrow music genres had a lower preference for loud apparel luxury, unless they had a greater need for status. Those liking lowbrow music genres did have a greater preference for loud apparel luxury, but this was mostly the case for higher-income consumers. In conclusion, the music genres people prefer to listen to are strongly correlated to their preference for loud or quiet apparel luxury. However, the need for status and the level of appreciation of diverse music styles moderate the valence of this association.

3

A

CKNOWLEDGEMENTS

I would like to thank first and foremost my supervisor Tina Dudenhöffer who has given me throughout this thesis process, the tools I needed to get the best out of this learning experience. By saying not too much, but just enough to make my brains sweat, she has been a crucial factor in my intellectual development during my master course.

In addition, I would like to thank Roger Pruppers for his enlightening ideas about the drivers behind consumer behavior. This has significantly increased my interest in the topic and thus my motivation for this thesis. Furthermore, I would like to thank Jorge Labadie for his willingness to stay longer after class to help me see the connections I could not see by myself.

Lastly, but not least. I would like to thank all the respondents who have taking the time to fill in my survey and send it around.

4

T

ABLE OF

C

ONTENTS

1.1 The next chapters ... 10

2 Literature Review ... 11

2.1 Status seeking behavior... 11

2.2 Luxury goods ... 12

2.3 Low-income consumers ... 13

2.4 Musical taste ... 15

2.5 Experiential Marketing ... 16

2.6 Theoretical framework ... 17

3 Data & Method ... 21

3.1 Participants ... 21 3.2 Measurement Variables ... 22 3.2.1 Independent variable ... 22 3.2.2 Dependent variable ... 22 3.2.3 Moderating variables ... 23 4 Results ... 24 4.1 Descriptive Statistics... 24 4.1.1 Omnivorousness ... 24 4.1.2 SUMHighbrow vs. SUMLowbrow ... 25 4.1.3 NeedForStatus ... 25 4.1.4 PrefLoudLuxury ... 26

5

4.2 Inferential Statistics ... 26

4.2.1 H1: Lowbrow Music Genre  Preference Loud Apparel Luxury ... 26

4.2.2 H2: Highbrow Music Genre  Preference Loud Apparel Luxury ... 28

4.2.3 H3a: Musical Omnivorousness  Lowbrow Music Genre – Preference Loud Apparel Luxury .. 29

4.2.4 H3b: Musical Omnivorousness  Highbrow Music Genre – Preference Loud Apparel Luxury . 30 4.2.5 H4a: Need For Status  Lowbrow Music Genre – Preference Loud Apparel Luxury ... 32

4.2.6 H4b: Need For Status  Highbrow music genre – Preference Loud Apparel Luxury ... 33

4.2.7 Multivariate Regression ... 34

4.3 Most Relevant Results & Limitations ... 37

5 Discussion & Conclusions ... 40

5.1 Implications ... 42

6 Appendix 1: Measurement Scales ... 44

7 Appendix 2: Transscript Focus Group ... 45

6

L

IST OF

F

IGURES

Figure 1- Skewness PrefMusicGenre ... 24

Figure 2- Preference Lowbrow vs. Highbrow Music Genres ... 25

Figure 3- Preference for Loud Apparel Luxury ... 26

Figure 4- Relation Lowbrow Music Genre & Preference Loud Apparel Luxury ... 26

Figure 5- Relation Highbrow Music Genres & Preference Loud Apparel Luxury ... 28

Figure 6- Preference lowbrow music genres and loud apparel luxury by gender ... 36

L

IST OF

T

ABLES

Table 2- Descriptive Statistics ... 24

Table 3- Nonlinear Regression Highbrow Music Genres ... 28

Table 4- Moderating Effect Musical Omnivorousness- lowbrow ... 29

Table 5- Musical Omnivorousness- Highbrow ... 31

Table 6- Moderating effect NeedForStatus- Highbrow ... 33

Table 7- Coefficients Multivariate Regression... 35

Table 8- Status Consumption Scale Factor Loadings ... 44

7

1 I

NTRODUCTION

“Luxury... corrupts at once rich and poor, the rich by possession and the poor by covetousness” – Jean-Jacques Rousseau.

It has been long observed that the rich and elite people consume luxury goods that reaffirm their position in the society while those who belong in the lower levels of society aspire to consume these products and services as well (Üstüner & Holt, 2010). Keeping this in mind, it is no surprise that luxury products are continuously described as the branded goods of which the mere use or display brings prestige to the owner, apart from any functional utility (Grossman & Shapiro, 1988). Often, this desire for luxury goods and services leads to lower class levels emulating the rich people’s consumption behavior to obtain status - an individual’s prestige position in relation to others. Low-income consumers are willing to pay premium prices for branded products if it signals a higher status rather than paying less for equivalent non-labeled products. Accordingly, the luxury market is significantly driven by the desire for status (Nelissen & Meijers, 2011). This is in line with recent findings concluding that the luxury market is not only driven by the rich, but by many less wealthy consumers (Plank & Cervantes, 2015). The personal luxury goods are hereby considered the “core of the core” of luxury and accounts for more than half of the luxury market. The personal luxury goods can be divided in the segments apparel, accessories, hard luxury (jewelry), fragrances & cosmetics, and others (Technavio, 2015). For instance, a Chanel coat is a luxury apparel item, while a Rolex watch belongs to the hard luxury segment.

How do people know what is luxurious and of high status? The media plays an important role in making people aware of what is considered luxury and what not. Media depictions of success are strongly correlated with the increase in the desire for status and luxury products. Media portrayals of a luxury life and stories of success are thought to create sketching images in consumers’ minds biasing their views of reality (Mandel, et al., 2006). This in turn impacts one’s expectations and his or her consumption preferences (Mandel, et al., 2006). An interesting note is that lower income households seem to be more exposed to media portrayals as they spend much more time with multimedia devices

8 than do higher-income households (Nielsen, 2015). Subsequently, the question that arises is in what way do media portrayals become as effective as influencing people’s desires and preferences? Several studies have determined that music plays a significant role in the effectiveness of media messages and often determines the affective response of message receivers (Seidman, 1981). This affective response in turn generates positive anticipated emotions that increases the desire for whatever the media message is selling (Hunter, 2006).

As music can evoke very strong feelings among consumers, many businesses have taking an interest in this matter (Habibi & Damasio, 2014). Experiential marketing1 _{is a worldwide growing trend in which}

music plays a major role (Atwal & Williams, 2008). In the luxury market, music is used in different ways. This ranges from music artists being associated with a brand to the creation of specific songs for the introduction of new products. Surprisingly, little is known about the role music plays in the desire for luxury products and how different people respond to different types of music. This thesis addresses this issue by taking a closer look at the affiliation between the preference for certain music genres and the preference for luxury products. There have been studies that focus on the music genres people of different social classes listen to. A distinction has been made between the popular, easy listening music genres termed lowbrow music genres, and the music genres that the elite, high class people listen to - highbrow music genres (Veenstra, 2015). Considering that the difference in social classes has an impact on the level of desire for status and luxury products, the distinction between lowbrow and highbrow music genres is applied throughout this thesis.

“Extravagance is the luxury of the poor; penury is the luxury of the rich” – Oscar Wilde.

When it comes to the type of luxury products people prefer, different studies have shown a strong relationship between the income level and the preference for loud or quiet luxury. People with a low-income have a higher preference for loud, obvious, luxury goods that have prominent logos and badges. For instance, a Gucci bag with the typical Gucci monogramming and the flashy logo can be

1 _{Marketing initiatives that give consumers in-depth, tangible experiences in order to provide them with}

9 considered loud luxury. On the other hand, people of greater wealth prefer quiet, discreet luxury goods with discrete logos and badges (Han et al. 2010). For instance, a Louis Vuitton t-shirt in which the logo only appears on the inside label. Loud luxury is often preferred by consumers who have a greater need for status and who try to emulate those whom they recognize as being wealthy (Han, et al., 2010). Low-income consumers have a greater need for status as the possession gap - The difference between what they have and what others have (Ordabayeva & Chandon, 2010) – is greater. These consumers often feel the need to consume more or focus more on luxury goods to achieve greater status and the accompanied position gains (the increase in social rank). As the luxury market is significantly driven by the need for status and low-income consumers are the people who are in greater need for status, it was expected that analyzing this group would ensure relevant data that could be used to make further inferences on the role music plays in the preference for luxury products. As low-income consumers, do not always have the financial resources to afford the expensive luxury goods, counterfeit products are also consumed with the objective to gain greater social status (Kempen, 2003). Taking this into consideration, in this study the need for status is represented by the tendency to purchase goods and services for their prestige as well as the appreciation for counterfeit status products. Furthermore, since loud luxury is said to be preferred by the status-seeking consumers, the focus of this study is the association of music genres and the preference for loud luxury. Within the loud luxury sector, the segment apparel represents the goods that are the most directly visible to others. This greater visibility was thought to stimulate the emulating behavior of people and thus the greater preference for loud luxury. Therefore, the emphasis laid on this segment.

Analyzing the music genres that stimulate a higher preference for loud luxury products can help luxury marketers determine in what way music can be used to target this status-seeking target market. Considering that the luxury market accounts for trillions (1.08 trillion Euros in 2016) (Bain & Company, 2016), researching this matter can contribute in a significant way and help gain new insights on how to enhance experiential marketing in this market. Taking all the factors into account, the main research question of this thesis is as follows:

10 What is the relationship between the preference for a music genre and the preference for loud luxury

apparel products by low-income consumers?

The premise here is that a greater need for status leads to a greater preference for loud apparel luxury. Furthermore, musical omnivorousness - the appreciation for diverse music styles – is expected to have a negative effect on the preference for loud luxury products. Omnivorousness is a means that represents open-mindedness which in turn reflects a higher educational level. A higher educational level is correlated with an improved self-esteem (Ma, et al., 2016) and thus a lesser need to gain a social status. Keeping this in mind, musical omnivorousness represents a more developed self-esteem and thus a lesser need to signal social status.

A low-income has long been linked to a lower educational level and several researchers have concluded that the lower educational level results in a greater appreciation for easy, relaxing, lowbrow arts (Claessens & Droest, 2010). Consequently, lowbrow musical taste is expected to have a positive effect on the preference for loud luxury products as it is expected that the appreciation for popular music could be a sign of trying to fit in and be part of a social group people consider beneficial. In this case, the social group would be the one with higher status. Highbrow musical taste is expected to have a negative effect on the preference for loud luxury apparel products since listening to highbrow music is associated with being of high class and commonly a lesser need for status. This lesser need for status is said to have a positive effect on the preference for quiet luxury rather than loud luxury.

1.1 T

In chapter 2, previous studies are evaluated and relevant ideas are discussed further. From here, areas that are still in need of further clarification are to be examined and a conceptual framework is presented. In chapter 3, the methodology used for this study is described, outlining the relevant variables and the sample used. In chapter 4, the descriptive results and the inferential results are discussed in detail as to identify whether the expected relations occurred. And lastly, but not least, in chapter 5 a thorough discussion is provided on the results obtained and what implications can be concluded from these findings.

11

2 L

ITERATURE

R

EVIEW

In this section, existing research is evaluated and discussed as to identify the areas that need further clarification.

2.1 S

Maslow once hypothesized that people follow a hierarchy of needs and that they feel the need to move up the hierarchy toward a level of self-esteem and self-actualization. Simply put, human beings will try to satisfy “new” needs whenever the existing ones are fulfilled. First, humans try to fulfill their physiological needs and from there on the more psychological needs come into play. One psychological category of needs that people strive to fulfill is self-esteem. Self-esteem is generally defined as a favorable opinion about oneself and how individuals feel about their own achievements (Vazir & Raza, 2009). William James (1910) once posited that all three aspects of the self - the material, the social and the spiritual - are capable of evoking feelings of happiness (Lyubomirsky, et al., 2006). Ironically, the self-esteem of a person and thus his or her opinion about own achievements is often relative and determined by the achievements of others (Solnick & Hemenway, 1998). This observation is explained by the social comparison theory that argues that people have the tendency to compare themselves to others to judge themselves (Festinger, 1954). One very important influencer of a person’s self-esteem is a person’s relative economic status (Carlsson, et al., 2003). Status is here defined as the social beliefs about an individual’s relative unobservable income (Moav & Neeman, 2010). In other words, status refers to an individual’s prestige position in relation to others. It has been established that individuals care a lot about their status and strive to attain higher status.

People with higher incomes consume products and services that reaffirm their position in the society and indicate their economic status. Those who belong in the lower levels of society aspire to consume these products and services as well to increase their status (Goodwin, et al., 2013). The “trickle down” theory has been used to explain this behavior and relates to the act of people in lower class levels emulating the people of higher class levels (Üstüner & Holt, 2010). This is where conspicuous consumption comes into play: people acquire goods or services to signal a social status (Veblen, 1899)

12 (Trigg, 2001) (O'Cass & McEwen, 2004). The most puzzling about conspicuous consumption is that those earning the least are also the ones that spend the greatest fraction of their income on conspicuous consumption. A possible explanation for this paradox is an individual’s psychological motive to protect self-integrity (Sivanathan & Pettit, 2010). Some scientists have explained that because the poor have so little that any increase in social status is of great value (Goodwin, et al., 2013). Furthermore, some sociologists believe that the social status is of great importance to the less fortunate as it might help these individuals gain access to the social benefits and rewards that normally are reserved for the elite (Banerjee & Duflo, 2007).

Another indicator of the need for status, is the appreciation of counterfeit positional goods. As low-income consumers, do not always have the financial resources to afford the more expensive positional goods, counterfeit products are often consumed or other deceptive status signaling strategies are conducted (i.e. abandoning positional goods discretely before having to pay) (Kempen, 2003). Low-income consumers have a greater need for status as the possession gap - The difference between what they have and what others have (Ordabayeva & Chandon, 2010) – is greater. Low-income consumers often feel that they need to consume more or focus more on luxury goods to achieve greater position gains (the increase in social rank). Taking all the previously mentioned factors into account, in this study the need for status is represented by the tendency to purchase goods and services for their prestige as well as the appreciation for counterfeit products.

2.2 L

Psychological research has continuously confirmed that the desire for status has been a significant driver for the luxury goods market (Nelissen & Meijers, 2011). This is in line with recent findings concluding that the luxury market is not only driven by the rich, but by many less wealthy consumers who aspire to gain the social benefits of luxury (Plank & Cervantes, 2015). Traditionally, luxury products are described as those branded goods of which the mere use or display brings prestige to the owner, apart from any functional utility (Grossman & Shapiro, 1988). Of the total luxury market, 80% is led by luxury cars, luxury hospitality and personal luxury goods (Bain & Company, 2016). The personal luxury

13 goods are hereby considered the “core of the core” of luxury and accounts for than half of the of the market. The personal luxury goods can be divided in the segments apparel, accessories, hard luxury, fragrances & cosmetics, and others (Technavio, 2015). Taking this into account, this research will focus on the segment apparel as it represents the goods that are the most directly visible to others. This greater visibility is thought to develop reactions from people in a faster and more measurable manner. One of these reactions is the emulating behavior of people.

Many luxury brands mark their products differently to be more or less visible in order to attract different types of consumers. This idea is explained by the construct of brand prominence. Brand prominence is the extent to which a product has visible markings that help ensure observers recognize the brand (Han, et al., 2010). A brand can choose to use ‘loud’ (obvious) branding or ‘quiet’ (discreet) branding for its products. The relative conspicuousness of a brand’s mark or logo is thought to reflect the different signaling intentions of the owner. This in turn has led several researchers to differentiate consumers by the way they signal status and to whom (Han, et al., 2010). Old-money consumers or ultra-rich consumers who have no real need for status seem to prefer less flashy luxury products that are only noticeable to those in their social group (i.e. with the use of smaller logos on their luxury item or no logos at all) (Husic & Civic, 2009). While consumers who have a strong need for status, lower-income consumers and nouveau rich consumers, tend to prefer loud luxury goods to signal a higher status and to emulate those who they recognize as being wealthy (Han, et al., 2010). As low-income consumers have a stronger need for status, it is expected that they would prefer loud luxury products over quiet luxury products.

2.3 L

-

In the European Union, poverty is defined as severe restrictions on material, cultural and social resources that exclude people from minimal participation (Scheeper & Grotenhuis, 2005). However, the EU-countries differ in the poverty threshold benchmarked in regards to their national income. Considering the 40%, 50% and 60% threshold of national median income, The Netherlands (11.6%), Denmark (11.9%), Slovakia (12.6%), Finland (12.8) and the Czech Republic (9.7%) are the countries with

14 the lowest poverty rates (Eurostat, 2015). It would be more interesting to look deeper into the spending behavior of the bottom-tier social classes in a country with low poverty rates as the greater wealth in the country might enable a greater social pressure. A greater social pressure is expected to stimulate the desire for a higher status among people in a society. Therefore, a closer look will be taken at the spending behavior of the bottom-tier class in the Netherlands.

In the Netherlands, poor consumers are those who earn equal or less than the poverty line income.

Table 1- Income classes the Netherlands

Net amount earned per month

Poor people < € 890 (= $950)

Low-income people € 890 ≤ €1511 (= $1612)

Average-income people €1720 (= $1835)

The governmental institution, the Social-Cultural Planning Office (SCP), has set the poverty line at €890 net a month per household (SCP, 2016). A low-income is considered an income equal to or less than €1511 net (= $1612) a month per household (Central Planning Office, 2016). The average income in the Netherlands on the other hand equals €1720 net (= $1835) a month (Central Planning Office, 2016). In order to meet the low-income people halfway, the Dutch government provides diverse allowances such as social security allowance, child support, and rent allowance only to those who are in the low-income group which include both poor and low-low-income people (Belastingdienst, 2017). Keeping this in might, a way to reduce bias when testing whether a person is part of the low-income group, the question whether they qualify for any allowances is to be presented.

It has long been observed that children from low-income families become low-income adults. Generational cycles of low income are quite common across countries and also in the Netherlands (Corak, 2006). Therefore, it is expected that low-income consumers often will come from a low-income family. Furthermore, coming from a low-income family might have a more positive effect on the need for status as having low-income has been the case for these people for a longer period of time and a need to escape this low status might be greater.

15

2.4 M

Several studies have determined that music plays a significant role in the effectiveness of media messages and often determines the affective response of message receivers (Seidman, 1981). This affective response in turn generates positive anticipated emotions that increases the desire for whatever the media message is selling (Hunter, 2006). In short, music is capable of triggering drives and emotions among its listeners (Habibi & Damasio, 2014). In everyday life, music is often used by people as a regulator of emotions and moods (Saarikallio & Erkkilä, 2007). Knowing the effect music can have on people’s behavior and attitudes, numerous researches have been conducted on the role that music plays in people’s life and whether differentiation among listeners can be observed. One such differentiation is the distinction of highbrow, middlebrow and lowbrow. The terms highbrow and lowbrow first appeared in the first quarter of the 20th _{century and refer to the connotation of the}

culture of the upper class and the culture of the mass respectively (Gans, 1974). Then the term ‘middlebrow’ was introduced to refer to the culture of those people who are neither highbrow or lowbrow, but aspire to be part of the high culture by showing an interest in highbrow culture (Lynes, 1949).

Cultural omnivorousness was introduced by Richard Peterson in the 90’s and refers to a greater breadth of cultural taste and a willingness to cross established hierarchical cultural genre boundaries (Peterson & Kern, 1996). Throughout the years, the openness toward both highbrow and lowbrow cultural tastes has become even more appreciated and seen as a source of social and cultural capital (Friedman, 2012). Omnivores are hereby defined as highbrow, high educated persons that appreciate not only highbrow art forms, but also middle- and lowbrow arts. When it comes to music as an art form, musical omnivorousness refers to the appreciation of diverse musical styles by highbrow persons. Furthermore, omnivores are generally people who are very self-confident and have a lesser need to prove others what their status is in society (Chan, 2013). A low-income has long been linked to a lower educational level and several researchers have concluded that the lower educational level results in a greater appreciation for easy, relaxing, lowbrow arts (Claessens & Droest, 2010). However,

16 with the rapid growth of information technologies and the access to an immense amount of data, people have become more informed and intelligent than ever before (Wineman, 2013). It is therefore expected that low-income consumers are also capable of engaging in omnivorousness and thus musical omnivorousness.

What kind of music is considered highbrow, middlebrow or lowbrow? Most studies have concluded a clear distinction between highbrow musical taste and everything that is not highbrow musical taste. For the sake of simplicity, throughout this thesis the distinction will be between highbrow and lowbrow musical taste. In regard to musical taste, highbrow musical taste refers to the appreciation for music genres that are more sophisticated and serious. On the other hand, lowbrow musical taste refers to the appreciation for music genres that do not demand much effort to understand and are easy to listen to (Veenstra, 2015). Typical highbrow music genres include blues, choral, classical, jazz, musical theater, opera, pop, reggae, rock, and world/international (i.e. Latin, African, Arabic) (Veenstra, 2015). Common lowbrow musical genres are country, disco, easy listening, golden oldies, heavy metal, and rap (Veenstra, 2015). When looking at the most-listened-to music genres in the Netherlands, the common genres are: Dutch pop, Songs about Life, Electronic dance, Opera, Nederhop (Dutch hip hop), Rap, Classic, Folk, Jazz, Rock, Heavy metal, Country, International/world, and R&B/Soul (Bloemendaal, 2000) (Gemeente Eindhoven, 2016). Some of the highbrow and lowbrow genres found by Veenstra are not significantly listened to in the Netherlands and will therefore not be used throughout the research. These genres are: easy listening, choral, and golden oldies.

Exclusive highbrow musical preference or exclusive lowbrow musical preference is very rare as some music genres are considered complementary to one another (Chan, 2013). Therefore, a combination of both is expected.

2.5 E

M

In the modern society, the marketing of luxury goods is not only associated with conveying an image of quality, performance and authenticity, but also with providing potential customers with an overall

17 experience (Atwal & Williams, 2008). Since the core of luxury goods is not its functionality, but rather the image it conveys and the prestige owners of luxury feel, experiential marketing is extremely powerful in the luxury goods market. So, what is experiential marketing? It refers to the amplification of a product’s essence into a set of tangible, physical and interactive experiences that reinforce the offering of this product (Atwal & Williams, 2008). A distinction is made among five different experiences marketers can provide: 1. sensory experiences (Sense), 2. affective experiences (Feel), 3. creative cognitive experiences (Think), 4. Physical experiences, behaviors & lifestyles (Act), 5. Social-identity experiences (Relate) (Schmitt, 1999). In general, the marketing of today contains aspects of all 5 experiences. When we look at the sensory experiences, consumers are lured by appealing to the senses of sight, sound, touch, taste and smell (Schmitt, 1999). On the other hand, affective experiences are achieved by triggering customer’s feelings and emotions. Recalling the significant impact music has on people’s emotions, moods and attitudes, it is no surprise that music is commonly used by luxury marketers to target their customers’ and offer a full product experience. However, what type of music triggers people to purchase luxury? Little attention has been given on the effect different music types have on people’s desire for luxury. This thesis addresses this issue by taking a closer look at the effect of music genres on people’s preference for luxury products. Considering that the luxury market accounts for trillions (1.08 trillion Euros in 2016) (Bain & Company, 2016), researching this matter can contribute in a significant way and help gain new insights on how to enhance experiential marketing in this market.

2.6 T

As mentioned previously, the luxury market is driven by people’s desire to gain a higher social status. Considering that low-income consumers have a stronger need for status, it would be beneficial to research how their preference for a music genre affects their desire for luxury. The importance placed on obtaining status is correlated with a higher preference for loud luxury products. Analyzing the music types that stimulates a higher preference for loud luxury products can help luxury marketers determine in what way music can be used to target the status-seeking consumer market. Therefore, the focus of

18

Figure 1- Conceptual framework: Preference music genre affects the preference for loud luxury apparel products.

this research lies on the preference for loud luxury products as it is assumed that low-income consumers will prefer loud luxury products over quiet luxury products. As apparel products is the most visible to people and would more easily lead to the emulating behavior that is often the case when seeking status, this study will focus on the loud luxury apparel segment.

A lower income is strongly correlated with a higher need for status. Based on this notion, it is expected that low-income consumers in the Netherlands have a higher need for status. Based on this assumption, the gap in the literature and the need for clarification on the role music plays in the desire for luxury products, a conceptual framework is designed.

The focus of this thesis is to determine the relationship between the preference for a music genre and the preference for loud luxury apparel products. It is expected that the preference for lowbrow music genres will have a positive effect on the preference for loud luxury products.

Therefore, the first hypothesis that will be tested is as follow:

H1: The preference for lowbrow music genres by low-income consumers has a positive effect on the

preference for loud luxury apparel products. (H1) (H4b) (H2) (H4a) (H3a) (H3b)

19 In contrast, the preference for highbrow music genres are expected to indicate a lower need for signaling status. This lower need for signaling status is expected to go hand in hand with a lower preference for loud luxury products. Therefore, the second hypothesis is as follows:

H2: The preference for highbrow music genres by low-income consumers has a negative effect on the

preference for loud luxury apparel products.

The relationship between the preference for a music genre and the preference for loud luxury products is thought to be modified by the omnivorousness of the low-income consumers in terms of musical taste. Therefore, the third hypothesis is explained with two sub hypotheses:

H3a: The positive relationship between preference for lowbrow music genres and the preference for

loud luxury apparel products is moderated by musical omnivorousness, so that this relationship is weaker for higher values of musical omnivorousness.

On the other hand, the negative relationship between the preference for highbrow music genres and the preference for loud luxury products is thought to be strengthened.

H3b: The negative relationship between preference for highbrow music genres and the preference for

loud luxury apparel products is moderated by musical omnivorousness, so that the relationship is stronger for higher values of musical omnivorousness.

As a greater need for status is linked to a greater preference for loud luxury, it is expected that the degree people need status moderates the relationship between the preference for a music genre and the preference for loud luxury apparel products.

H4a: The positive relationship between the preference for lowbrow music genres and the preference

for loud luxury apparel products is moderated by the need for status, so that the relationship is strengthened for higher values of need for status.

However, the negative relationship between the preference for highbrow music genres and the preference for loud luxury apparel products is expected to be negatively moderated.

20 H4b: The negative relationship between the preference for highbrow music genres and the preference

for loud luxury apparel products is moderated by need for status, so that the relationship is weakened for higher values of need for status.

21

3 D

ATA

&

M

ETHOD

In order to answer the main research question of this thesis: ‘What relationship is there between the preference for a music genre and the preference for loud luxury apparel products by low-income

consumers?’, an experimental survey was conducted. In this section of the thesis, the methods and

procedures to test the aforementioned hypotheses are outlined.

3.1 P

The focus group of this study was low-income consumers and consequently different private Facebook groups in which presumably low-income consumers could be found, were approached. The main objective of these private groups was to build a community in which the members give each other suggestions on how to best handle their income and where to get the best product deals. A total of 361 people participated in this study of which 43 persons eventually did not submit the complete survey and this data could not be used for further inferences. To check for missing data from the surveys that were submitted, frequency tests were conducted on the remaining sample of 318 participants and from the results 24,5% represented missing data. The 78 cases that had missing data were excluded from the analysis. The final sample that was used for further analyses comprised of 240 participants with 85,4% female and 14,6% male. Furthermore, the greatest number of respondents 37,1% were within the age group of 25 till 34 years old and with 74,9% being older than 25 years. None of the participants were older than 74 years old.

A good representation of low-income consumers and poor consumers was obtained with a 64,6% of the participants earning equal to less than €1511. In addition, 27,1% of the participants had parents with a low income and 34,6% has parents that earned an average income. To test whether people were answering the question about their income level honestly, the respondents were asked to indicate whether they received any governmental allowances. Of the 240 respondents, 60,4% indicated to receive some type of governmental allowance. When we look at the educational level of the

22 participants, the majority did not enjoy a high education: 67,9% followed only high school or an additional intermediate vocational education (MBO).

3.2 M

V

Following Veenstra’s genres for highbrow and lowbrow musical taste, the respondents were asked to indicate which of the following genres they prefer to listen: classical music, hip hop, folk music, rap, opera, country music, pop, jazz, reggae, rock, heavy metal, blues, world/international, disco, electronic dance. Integrating both the typical highbrow music genres and the most-listened-to highbrow music genres in the Netherlands, the highbrow music genres presented to the respondents were: blues, classical, jazz, opera, pop (including Dutch pop), reggae, rock, folk, and world/international (including Latin, African, Caribbean & Eastern). The highbrow measure was a nominal variable ranging from 0 till 9. The lowbrow genres that were presented to respondents are: songs about Life (smartlappen), electronic dance, Nederhop (Dutch hip hop), rap, country, R&B/soul, disco, and heavy metal (Veenstra, 2015). The lowbrow measure was a nominal variable ranging from 0 till 9. The overall preference for different music genres was the sum of both the lowbrow music genres and the highbrow music genres. This measure was a ratio variable ranging from 0 till 18.

3.2.2 Dependent variable Preference Loud luxury

As a way to test for the preference for loud apparel luxury, 2 status apparel products were presented to the respondents after which they had to indicate which one they liked best. Women and men were exposed to different items. Furthermore, the respondents got a total of four questions in which two items – one with discreet labeling and one with obvious labeling – were presented. This measure was represented by a dichotomous variable. In order to identify the items which best to use a pretest was conducted by having a focus group of 7 persons. The focus group was not structured as the objective was to get a real glimpse of what people’s thoughts were in regard to what they considered quiet and

23 loud luxury. The focus laid here on whether the brand was easy to distinguish because of its conspicuousness. Furthermore, their opinions on the brands to use, the products to use and the best level of similarity between products were obtained and incorporated in the survey. Since the focus sample was low-income consumers, low-income consumers were also approached for the pretest.

3.2.3 Moderating variables Need for status

As mentioned in the previous section, the need for status is represented by the tendency to purchase goods and services for their prestige as well as the appreciation for counterfeit products. To test for this construct two scales were used: The Status Consumption Scale and the Attitude Toward Counterfeited Products Scale. The 5-item Status Consumption Scale with factor loadings ranging from 0.59 to 0.83 significantly measured the extent to which the respondents buy or use products for the status they confer (Eastman et al. 2017). The respondents were asked to indicate to which extent they agreed with each item presented using a 5 point Likert-scale (1 = strongly disagree to 5 = strongly agree). The 5-item Attitude Toward Counterfeited Products Scale had a Cronbach’s alpha of 0.85 and measured the participant’s attitude towards counterfeited products using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree)(Alberto et al. 2007). Both scales were represented by an ordinal variable. The need for status was the sum of both variables and was represented by an interval variable. Please refer to appendix 1 for both scales and its items.

Musical omnivorousness

To test for the musical omnivorousness, the respondents were given the opportunity to select more than one music genre. Musical omnivorousness occurred whenever the respondents selected multiple music genres. This measure was a ratio variable ranging from 0 till 18 as it included both highbrow and lowbrow music genres.

24

4 R

ESULTS

In this section of the thesis, the results of the overall study are provided. A combination of descriptive and inferential analyses was conducted and are further discussed in the following pages.

4.1 D

S

Before being able to analyze the data, several variables needed to be merged, recoded or deleted to ensure that any future inferences are based on adequate data. A total of 12 outliers were found, but none of these significantly affected the mean when looking at the 95% trimmed mean and thus were not extreme. To avoid losing relevant data these outliers were kept in the dataset. Below is an overview of the final variables that represent the constructs in the conceptual model of chapter 2 and their associated descriptive statistics.

Table 2- Descriptive Statistics

Table 2: Descriptive Statistics variables

Min Max Mean Median Std. Dev. Variance Skewness Kurtosis

Omnivorousness 1 13 3.863 4.000 2.199 4.839 1.154 1.976 SUMHighbrow 0 8 2.146 2.000 1.552 2.410 1.196 1.961 SUMLowbrow 0 5 1.717 2.000 1.060 1.124 0.226 0.020 NeedForStatus 0 9 2.229 2.000 1.859 3.458 1.161 1.199 PrefLoudLuxury 0 4 1.912 2.000 1.029 1.059 -0.102 -0.583 Number of observations 240 4.1.1 Omnivorousness

This variable represents the number of music genres that the participants preferred to listen to and includes both lowbrow and highbrow music genres. All responses range from 1 till 13 with a mean average of 3.863, a median of 4.000 and a standard deviation of 2.199. Considering the maximum value of 13, the average of 3.863 is a rather low value and it seems that the participants have a quite narrow musical taste. When we look at the shape of the data for this variable, it appears to be highly skewed to the right even though the median is greater than the mean. This skewness to the right seems

Figure 1- Skewness PrefMusicGenre

Preference Music Genres

Fr e q u e n cy

25 to support the idea that the majority of the respondents had a narrow musical taste and that those who appreciated a greater number of diverse musical styles reflected the minority. In other words, musical omnivorousness was not reflected by the majority. Furthermore, given the somewhat high value of the kurtosis, the distribution is more clustered and the shape has a leptokurtic form resulting in a higher peak around the lower values.

4.1.2 SUMHighbrow vs. SUMLowbrow

These variables measure the number of music genres that are lowbrow and that are highbrow. For SUMHighbrow, the initial range of responses was from 0 till 9 and from the responses obtained the range is from 0 till 8 with a mean of 2.146, a median of 2.000 and a standard deviation of 1.552. The data seems to be skewed to the right and which explains why the mean is larger than the median. When looking at SUMLowbrow, all responses range from 0 till 5 with a mean of 1.717, a standard deviation of 1.060 and a median of 2.000. Considering that the initial range for lowbrow genres was from 0 till 9, the maximum value of 5 in the latest range is rather surprising. From the given data, it seems that the participants favored more the highbrow music genres than the lowbrow music genres. Since the majority of the respondents did have a low-income and as mentioned before, low-income consumers tend to prefer easy, relax lowbrow arts, it is astonishing that there is a greater preference for highbrow music genres. In regard to the shape of distribution, the data is approximately symmetric.

4.1.3 NeedForStatus

This variable represents the level of need for status that the respondents have. This variable is a merge of the Status Consumption scale and the Attitude Towards Counterfeited Products scale. The responses for this variable range from 0 till 9 with a mean of 2.229, a median of 2.000 and a standard deviation of 1.859. When looking at the somewhat low mean, it appears that the majority of the respondents did not care much about obtaining status. Again, this is surprising when considering the

26 low-income issue. This variable is skewed to the right with the mean being greater than the median. In addition, the data has a leptokurtic form with a kurtosis of 1.199.

4.1.4 PrefLoudLuxury

This variable represents the level of preference for loud apparel luxury. It ranges from 0 till 4 with a mean of 1.912, a median of 2.000 and a standard deviation of 1.029. It seems that there is not a high nor a low preference for loud luxury and that there is a normal distribution of data as can also be seen in figure 3. In regard to the shape of distribution, the data is slightly left skewed but still rather symmetric with a skewness of -0.102.

4.2 I

S

In this section of the thesis, a closer look will be taken at the expected relationships visualized in the conceptual framework of chapter 2. Several analyses have been conducted with the objective to find empirical evidence for the hypotheses mentioned in the chapter 2.

4.2.1 H1: Lowbrow Music Genre  Preference Loud Apparel Luxury

The first hypothesis suggested that the preference for lowbrow music genres positively affects the preference for loud luxury apparel. To test the first

hypothesis a Pearson Correlation Test was conducted as a way to not only determine whether the relationship between the constructs is positive, but whether this relationship is significant as well. From the Pearson correlation test, an P-value of 0.012 was obtained with a significance of 0.857. The relationship between the preference for lowbrow music genres and the preference

Figure 3- Preference for Loud Apparel Luxury

Figure 4- Relation Lowbrow Music Genre & Preference Loud Apparel Luxury

27 for loud apparel luxury is slightly positive as expected. However, the relationship is not strong nor significant. As we can see from the graph in figure 4, there is a bend in the line. Since the relationship does not appear to be linear a non-linear regression was conducted by taking the square variant of SUMLowBrow. An R-Square value of 0.075 was obtained, showing that the new model was only predicting 7.5% of the variance in the outcomes of the preference for loud apparel luxury. Furthermore, a significance level of 0.256 was generated, so the relationship remained statistically insignificant and so it is not to say that the variance in the preference for loud apparel luxury is cause by the preference for lowbrow music genres. Furthermore, when looking at the low-income consumers alone, the effect remained insignificant. Nevertheless, it can be said that there is an increase in the preference for loud apparel luxury up till a moderate level of preference for lowbrow music genres. Additional preference for lowbrow music genres goes hand in hand with a decrease in the preference for loud apparel luxury. The bend in this relation might be because the majority of the respondents seemed to prefer more highbrow music genres.

However, when looking at the participants that did not have a low-income, an R-value of 0.185 and a significance of 0.090 was obtained. Taking the 90% confidence interval, this result is considered modestly and positively significant in this thesis. Since, this thesis aims to get some more understanding on the relationship between music genres and consumer behavior, a lesser uncertainty with enough strictness for the results to remain relevant, is accepted. From this result obtained, it seems that the respondents with higher incomes who prefer lowbrow music genres indeed have a greater preference for loud apparel music genres. This is a quite interesting result considering that people from lower incomes were expected to show a stronger relationship between the preference for lowbrow music genres and the preference for loud apparel luxury. The first hypothesis is therefore partially supported: preference for lowbrow music genres does have a positive effect on the preference for loud apparel luxury, but this effect is mostly apparent for people with higher incomes.

28 4.2.2 H2: Highbrow Music Genre  Preference Loud Apparel Luxury

The second hypothesis that is to be tested, posits that the preference for highbrow music genres negatively affects the preference for loud luxury apparel. As both variables are of quantitative character, a Pearson Correlation test was conducted. Having conducted the analysis, an R-value of - −0.019 was obtained with an accompanied significance of 0.659. There is a negative relationship between the preference for highbrow music

genres and the preference for loud apparel luxury, but this relationship is slight and not significant. However, when we look at figure 5, a clear pattern can be observed in which increasing values of highbrow music preferences are more than often associated with decreasing values of preference for loud apparel luxury. Taking this into consideration, a nonlinear regression was conducted in which the square of SUMHighbrow was utilized. Below are to find the results obtained.

Table 3- Nonlinear Regression Highbrow Music Genres

Model R Square Change Sig. Change B

1 0.001 0.659

SUMHighbrow -0.019

2 0.012 0.097

SUMHighbrow -0.198

Highbrow^2 0.029

As we can see in table 3, the overall model did improve when inserting the square variable of SUMHighbrow, but it remained statistically insignificant and only 1.2% predictive for the outcomes of the dependent variable. However, keeping in mind that a low-income was expected to show the negative correlation with the preference for loud apparel luxury, another analysis was conducting excluding those who did not have a low-income. From this analysis, a Pearson-correlation value of −0.200 was obtained with a significance value of 0.013. So, the relationship between the preference for highbrow music genres by low-income consumers did have a negative impact on the preference for loud apparel luxury and thus hypothesis 2 is supported by the obtained results.

Figure 5- Relation Highbrow Music Genres & Preference Loud Apparel Luxury

29 4.2.3 H3a: Musical Omnivorousness  Lowbrow Music Genre – Preference Loud Apparel Luxury The third hypothesis treats musical omnivorousness as a moderating effect in the relationship between the preference for lowbrow music genres and the preference for loud apparel loud luxury. Before being able to test the hypothesis, first the variables in question needed to me centralized and afterwards multiplied. In this way, a new variable – LowBrowMO - is created and is further used to test the moderation. Consequently, a regression analysis has been conducted. Below is to find the regression model used for this analysis.

𝑃𝑟𝑒𝑓𝐿𝑜𝑢𝑑𝐿𝑢𝑥𝑢𝑟𝑦 = 𝛼 + 𝛽1𝑆𝑈𝑀𝐿𝑜𝑤𝑏𝑟𝑜𝑤 + 𝛽2𝑀𝑂 + 𝛽3𝐿𝑜𝑤𝐵𝑟𝑜𝑤𝑀𝑂 + 𝜀, 𝜀~𝑛(0, 𝜎2) Where:

MO is the level of musical omnivorousness

LowBrowMO is the moderator – multiplication of SUMLowbrow and MO.

From the first analysis conducted, an R-Square value of 0.0308 with a significance of 0.842 was obtained for the model excluding the moderator. When including the moderating effect, an R-square value of 0.074 was obtained with a significance value of 0.729. Though the model improved somewhat, the model remained insignificant and unpredictive for the outcomes for the preference for loud apparel luxury.

Table 4- Moderating Effect Musical Omnivorousness- lowbrow

Variable Coeff. Std. Error t p

Low-income consumers

Constant 2.265 0.165 13.745 0.000 SUMLowbrow 0.109 0.120 0.911 0.364

Omnivorousness -0.132 0.061 -2.175 0.031

LowBrowMO 0.020 0.057 0.344 0.731

Modal- to high-income consumers

Constant 1.371 0.236 5.804 0 .000 SUMLowbrow 0.012 0.166 0.075 0.941 Omnivorousness 0.107 0.074 1.445 0.152 LowBrowMO -0.111 0.126 -0.880 0.381

30 When looking at the low-income consumers separately, relevant results were obtained. The model became statistically significant with an R-square of 0.201 and a significance of 0.044. When including the moderating effect, the model deprecated with an R-square of 0.203 and a significance of 0.096. Looking at the respondents that did not have a low-income, an R-square of 0.241 was obtained with a significance of 0.085. When including the moderating effect, the model became unpredictive for the outcomes of the dependent variable with an R-square of 0.259 and a significance of 0.129. Thus, there is no moderating effect through omnivorousness, but it does have an impact on the preference for loud apparel luxury when looking at the low-income consumers.

When looking at the effect size of the variables individually, SUMLowbrow remained insignificant on the preference for loud apparel luxury whether chosen by low-income consumers or those who had higher incomes. This might have to do again with the greater preference for highbrow music genres. Omnivorousness does seem to have a negative impact on the model when looking at low-income consumers alone. This moderation is going to be further researched when conducting a multivariate regression in order to see if other control variables haven an effect on the significance of the moderation.

4.2.4 H3b: Musical Omnivorousness  Highbrow Music Genre – Preference Loud Apparel Luxury The sub-third hypothesis suggested that the level of musical omnivorousness would moderate the relationship between the preference for highbrow music genres and the preference for loud apparel luxury in a way that would make the relation stronger. In order to test this hypothesis, the variables in question were first centralized and then multiplied with the objective to form a new variable, HighBrowMO. With this variables moderation was tested with the use of a regression analysis. The regression model for this test is created as follows:

𝑃𝑟𝑒𝑓𝐿𝑜𝑢𝑑𝐿𝑢𝑥𝑢𝑟𝑦 = 𝛼 + 𝛽1𝑆𝑈𝑀𝐻𝑖𝑔ℎ𝑏𝑟𝑜𝑤 + 𝛽2𝑀𝑂 + 𝛽3𝐻𝑖𝑔ℎ𝐵𝑟𝑜𝑤𝑀𝑂 + 𝜀, 𝜀~𝑛(0, 𝜎2) Where:

MO is the level of musical omnivorousness

31 The first regression analysis conducted provided an R-Square value of 0.038 and a regression significance of 0.842. When including the moderating effect the R-square obtained was 0.064 with a significance value of 0.805. Though the model improved somewhat, it remained insignificant in predicting the values of the preference for loud apparel luxury.

When looking at the low-income consumers, the model without the moderating effect was significant with an R-square of 0.201 and a significance of 0.044. Once the moderating effect was included, the model deprecated with a R-square of 0.204 and a significance of 0.093. The same occurred when looking at the respondents that did not have a low-income. The initial model had a R-square of 0.241 with a significance of 0.085. Once the moderating effect was included the model became insignificant with a R-square of 0.267 and a significance of 0.111. So, again omnivorousness does not represent a moderating effect.

Table 5- Musical Omnivorousness- Highbrow

Surprisingly, when looking at the parameters individually, none had a significant effect on the preference for loud apparel luxury. Since the models in table 5 did prove to be statistically significantly predicting the outcome of the preference for loud apparel luxury, it appears that SUMHighbrow and Omnivorousness are important predictor variables in this case. Musical Omnivorousness did not seem to moderate the relationship between the preference for highbrow music genres and the preference

Variable Coeff. Std. Error t p

Low-income consumers - model 1

Constant 2.265 0.165 13.745 0.000 SUMHighbrow -0.109 0.120 -0.911 0.364 Omnivorousness -0.023 0.081 -0.277 0.782

Modal- to high-income consumers – model 1

Constant 1.371 0.236 5.804 0 .000 SUMHighbrow 0.012 0.166 -0.075 0.941 Omnivorousness 0.107 0.074 1.445 0.152

32 for loud apparel luxury significantly. This moderation will be looking further into when conducting a multivariate regression analysis.

4.2.5 H4a: Need For Status  Lowbrow Music Genre – Preference Loud Apparel Luxury

The fourth hypothesis stated that the NeedForStatus would moderate the relation between the preference for lowbrow music genres and the preference for loud apparel luxury products in a way that the relationship becomes stronger for higher values of Need for Status. As NeedForStatus acts as a moderator in this relation, before being able to test the hypothesis, the variables in question needed to be centralized. Afterwards the variables were multiplied and a new variable – LowStatus - was designed to take into further consideration. A regression analysis was conducted. Below is to find the regression model designed beforehand.

𝑃𝑟𝑒𝑓𝐿𝑜𝑢𝑑𝐿𝑢𝑥𝑢𝑟𝑦 = 𝛼 + 𝛽1𝑆𝑈𝑀𝐿𝑜𝑤𝑏𝑟𝑜𝑤 + 𝛽2𝑁𝑒𝑒𝑑𝐹𝑜𝑟𝑆𝑡𝑎𝑡𝑢𝑠 + 𝛽3𝐿𝑜𝑤𝑆𝑡𝑎𝑡𝑢𝑠 + 𝜀, 𝜀~𝑛(0, 𝜎2) Where:

NeedForStatus is the level of need for status

LowStatus is the moderator – multiplication of SUMLowbrow and NeedForStatus.

The first model had an square of 0.007 and a significance of 0.418, while the second model had a R-square of 0.008 and a significance of 0.618. Given the results obtained from the regression analysis, the regression model does not statistically significantly predict the outcome of PrefLoudLuxury. The results remained insignificant when looking at the low-income consumers separately. However, when taking a closer look at the respondents with the higher incomes, at a 90% significance level, the model is significant with an R-square of 0.063 and a significance value of 0.071. Though only 6.3% of the variance can be explained by the model, this finding could be a first step into future research on this matter. The parameters individually did not have a significant effect on the preference for loud apparel luxury and also not when the moderator was included. Only SUMLowbrow and NeedForStatus taken together represented a significant model. The contribution that each variable had on the model on an individual level is shown below with the associated regression equation.

33 Here, both variables have a positive impact on the model. SUMLowbrow had a significance value of 0.113, NeedForstatus of 0.120. Given the results, no mediation is taking place. Furthermore, when testing for multicollinearity, a tolerance of 0.994 for both variables was obtained, illustrating that there was no multicollinearity between the variables. So, it seems that the preference for lowbrow music genres by higher income consumers only impacts the preference for loud apparel luxury when there is need for status. However, the results do not show a moderation effect. Taking this into account, hypothesis 4a can be rejected.

4.2.6 H4b: Need For Status  Highbrow music genre – Preference Loud Apparel Luxury

The sub-fourth hypothesis suggested that the relation between the preference for highbrow music genre and the preference for loud luxury is moderated by NeedForStatus in a way that the relation is weaker for higher values of NeedForStatus. Again, a moderating relation is to be tested and thus the variables in question were first centralized and afterwards multiplied. The new variable obtained was called HighStatus. Consequently, a regression analysis was conducted. Below is to find the regression model that was designed beforehand.

𝑃𝑟𝑒𝑓𝐿𝑜𝑢𝑑𝐿𝑢𝑥𝑢𝑟𝑦 = 𝛼 + 𝛽1𝑆𝑈𝑀𝐻𝑖𝑔ℎ𝑏𝑟𝑜𝑤 + 𝛽2𝑁𝑒𝑒𝑑𝐹𝑜𝑟𝑆𝑡𝑎𝑡𝑢𝑠 + 𝛽3𝐻𝑖𝑔ℎ𝑆𝑡𝑎𝑡𝑢𝑠 + 𝜀, 𝜀~𝑛(0, 𝜎2)

Where:

HighStatus is the moderator – multiplication of SUMHighbrow and NeedForStatus.

The initial regression analysis conducted provided an R- square value of 0.008 at a significance of 0.403. When we look at the second model, a

R-square of 0.024 at a significance of 0.123 was obtained. Though the model remained statistically insignificant, a closer look had to be taken at the factors affecting the model’s variance prediction to such an extent to see whether moderation was taking place. In table 6, the results obtained per parameter can be found.

Variable Coeff. Std. Error t p

Model 1 Constant 1.840 0.144 12.767 0.000 NeedForStatus 0.046 0.036 1.276 0.203 SUMHighbrow -0.014 0.043 -0.322 0.747 Model 2 Constant 1.856 0.143 12.938 0.000 NeedForStatus 0.036 0.036 0.989 0.323 SUMHighbrow -0.017 0.043 -0.389 0.698 HighStatus -0.136 0.068 -1.997 0.047 Table 6- Moderating effect NeedForStatus- Highbrow

34 As can be seen in table 6, the interaction effect is significant when included into the model, but the main effects- NeedForStatus and SUMHighbrow – remain insignificant. Given the results, it is to say that the relationship between preference for loud apparel luxury and the preference for highbrow music genres depends on the values of NeedForStatus. In addition, the moderation is negative, suggesting that the relationship between the preference for highbrow music genres is weakened by higher values of NeedForStatus. Keeping this in mind, hypothesis 4b is supported.

4.2.7 Multivariate Regression

In order to test whether the variables that are said to predict the preference for loud apparel luxury, do show significant results when taking as a whole, a multivariate regression is conducted. Furthermore, the multivariate regression as conducted with the objective to see whether other variables might have an effect on the preference for loud apparel luxury. Below is to find the regression model used for this analysis.

𝑃𝑟𝑒𝑓𝐿𝑜𝑢𝑑𝐿𝑢𝑥𝑢𝑟𝑦

= 𝛼 + 𝛽1𝑆𝑈𝑀𝐿𝑜𝑤𝑏𝑟𝑜𝑤 + 𝛽2𝑆𝑈𝑀𝐻𝑖𝑔ℎ𝑏𝑟𝑜𝑤 + 𝛽3𝑂𝑚𝑛𝑖𝑣𝑜𝑟𝑜𝑢𝑠𝑛𝑒𝑠𝑠

+ 𝛽4𝑁𝑒𝑒𝑑𝐹𝑜𝑟𝑆𝑡𝑎𝑡𝑢𝑠 + 𝛽5𝐺𝑜𝑣𝑒𝑟𝑛𝑚𝑒𝑛𝑡𝑎𝑙𝐴𝑙𝑙𝑜𝑤𝑎𝑛𝑐𝑒 + 𝛽6𝐴𝑔𝑒 + 𝛽7𝐺𝑒𝑛𝑑𝑒𝑟

+ 𝛽8𝐼𝑛𝑐𝑜𝑚𝑒 + 𝛽9𝐸𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛 + 𝛽10𝑃𝑎𝑟𝑒𝑛𝑡𝐼𝑛𝑐𝑜𝑚𝑒 + 𝛽11𝐻𝑖𝑔ℎ𝑆𝑡𝑎𝑡𝑢𝑠

+ 𝛽12𝐿𝑜𝑤𝑆𝑡𝑎𝑡𝑢𝑠 + 𝛽13𝐻𝑖𝑔ℎ𝐵𝑟𝑜𝑤𝑀𝑂 + 𝛽14𝐿𝑜𝑤𝐵𝑟𝑜𝑤𝑀𝑂 + 𝜀, 𝜀~𝑛(0, 𝜎2)

Having conducted the multivariate regression analysis, some interesting results were obtained. The overall regression model generated an R- Square value of 0.384 with a significance of 0.000. This indicates that the overall model though modest, is statistically significantly predicting the outcome of the preference for loud apparel luxury. The variable for the parent’s income had no significant effect on the preference for loud apparel luxury (r= -0.027 p=0.731) and seemed to lower the value of the model when running the regression with and without this variable. Therefore, this variable was taking out and when including the moderators, the model improved to an R-square of 0.425 and a significance of 0.000.

35

Table 7- Coefficients Multivariate Regression

When looking at the contribution of the variables individually, the variables that showed a significant effect are Gender, Education, HighBrowMO, and LowBrowMO. When excluding the non-significant variables in the model, the overall model deprecated to an R-square of 0.408 at a significance of 0.000. Thus, it seems that the moderators do play an important role in the model and to avoid losing relevant data, all variables are kept.

HighbrowMO

The HighBrowMO represents the moderating effect of omnivorousness on the relationship between the preference for highbrow music genres and the preference for loud apparel luxury. Interestingly, this moderator turns out to be significant in this multivariate model, while in the previous moderation analysis it did not appear to be significant. The moderator positively effects the regression model when controlling for the other variables in the model. Greater values of omnivorousness have a

36 strengthening effect on the negative relationship between the preference for highbrow music genres and the preference for loud apparel luxury.

Gender

Gender significantly affects the model in predicting the outcome of preference for loud apparel luxury. This is an interesting result at most of the respondents were female. As can be seen from figure 6, there is quite a difference in the preference for loud luxury apparel between men and women. Men seem to follow the expected trend: a greater preference for loud luxury apparel is associated with a decrease in the preference for highbrow music genres. However, for women the opposite occurs: a greater preference for loud apparel luxury is associated with a

greater preference for highbrow music genres. Furthermore, when looking at figure 9, it can be seen that in terms of the preference for lowbrow music genres, gain men and women differ significantly. Women follow the expected pattern in that a greater preference for loud apparel luxury goes hand in hand with a greater preference for lowbrow music

genres. On the other hand, men do not seem to follow this pattern and actually have a lesser preference for lowbrow music genres when having a greater preference for loud apparel luxury. It is important to note again, that the respondents were 85,4% female. Therefore, it is possible that a greater suitable balance between male and female might provide different results and thus further investigation is recommended.

Figure 6- Preference lowbrow music genres and loud apparel luxury by gender

Figure 7- Preference highbrow music and loud apparel luxury by gender