Consumer lifestyle and brand preferences : investigating the impact of consumer lifestyle on preferences towards fashion brands

(1)

Consumer lifestyle and brand preferences:

Investigating the impact of consumer lifestyle on preferences

towards fashion brands

Master Thesis – Business Administration – Marketing Track

Amsterdam, August 30, 2015 (final)

Student name: An Ly Nguyen Student number: 10001689 Faculty: Economics and Business Academic year: 2014 - 2015

Supervisor: Drs. Ing. A.C.J. Meulemans Second corrector: Prof. Dr. J.H.J.P. Tettero

(2)

Statement of originality

This document is written by Student An Ly Nguyen 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.

(3)

Abstract

Consumer lifestyle has been found to be a useful concept for marketing and advertising purpose, since the more brand managers and marketers know and understand about their customers, the more effectively they can communicate with and serve their customers. This study argues that consumer lifestyle influences brand preferences and the focus will be on preferences towards fashion brands with different brand personalities. Furthermore, this paper provides a greater understanding of the moderation effect of consumer characteristics on the relationship between consumer lifestyle and fashion brand preferences. The data was collected by self-administered questionnaires from 152 Dutch and international respondents. The findings of this study indicate that consumer lifestyle does influence consumers’ preferences towards fashion brands. Moreover, the findings suggest that consumer characteristics do not necessarily moderate the relationship between consumer lifestyle and fashion brand preferences. In other words, consumers with a particular lifestyle prefer to use fashion brands with a specific brand personality, however this relationship is not necessarily influenced by the personality traits that consumers possess.

(5)

1. Introduction

Lifestyle is increasingly important in the field of marketing communication. It has been shown that not only consumer personality, but also consumer lifestyle influences consumption patterns and consumer behavior (Vyncke, 2002). Lifestyle can be defined as the patterns in which people live and spend their time and money in order to differentiate themselves from others (Kaynak and Kara, 2001; Vyncke, 2002). The concept of lifestyle represents a set of ideas quite distinct from that of the Big Five consumer personalities (Kucukemiroglu, 1999), since the Big Five model describes a consumer’s personality by cognitive, emotional and behavioral elements (Mulyanegara & Tsarenko, 2009).

Consumers often prefer brands that are similar to their own personality (Huang, Mitchell & Rosenaum-Elliot, 2012). A brand personality can attract consumers’ attention such that the consumer may form a preference for the brand (Mulyanegara, Tsarenko & Anderson, 2009). Brand personality can be defined as “a set of human characteristics associated with a brand” (Aaker, 1997: 347). According to Aaker and Fournier (1995) a brand can function as a character, partner and person. Aaker (1997) therefore came up with the five different brand personality dimensions: Sincerity, Excitement, Competence, Sophistication and Ruggedness.

Prior research has shown that human personality could also influence the preferences of consumers towards particular brands (Huang et al., 2012). Different human personalities according to the Big Five model are: Neuroticism, Extroversion, Openness to Experiences, Agreeableness and Conscientiousness. Moreover, consumers’ demographics such as gender and education could also affect consumers’ brand preferences. For example Mulyanegara et al. (2009) suggest that male consumers are more self-expressive in their brand preferences compared to female consumers.

(6)

Previous studies on the lifestyle topic have mainly investigated the relationship between consumer lifestyle and consumption patterns. Moreover, previous studies about brand preferences are mainly focused on the relationship between consumer personality and brand preferences. In this study the focus is on the influence of consumer lifestyle on preferences towards fashion brands and if consumer characteristics influence this relationship. The effect of consumer lifestyle on preferences towards fashion brands has not yet been investigated thoroughly before. This study therefore aims to contribute to the lifestyle literature, branding literature, the consumer behavior literature, and to fashion research. This study can contribute to the knowledge about effective ways for fashion brands to communicate with and serve customers with a particular lifestyle.

In order to fill this gap in the literature this study will investigate the effect of consumer lifestyle on fashion brand preferences and if this relationship is influenced by consumer characteristics. The research question therefore will be: How does consumer

lifestyle influence preferences towards fashion brands and how is this relationship influenced by consumer characteristics? To provide an answer to this question, a survey questionnaire

will be held among Dutch and international consumers. The results of the survey can help brand managers in the fashion industry to determine how to make their brand more attractive for consumers and what kind of consumers they should target. According to Lin (2002) both lifestyle and consumer characteristics are effective variables for identifying market segments and targeting consumers.

This paper will follow up with a literature review, where prior studies are used to explore the topic. Thereafter, the conceptual model will be presented. Then, the methodology will be explained. Subsequently, the results will be presented and discussed. Finally, the paper will end with a conclusion and the answer on the research question.

(7)

2. Literature review

In this section the existing literature related to the research problem will be discussed in order to explore the research topic consumer lifestyle more precisely. First, the variables consumer lifestyle, brand preferences and consumer characteristics will be defined and explored. Thereafter, the literature review will end with a short conclusion, which emphasizes the need for this research about the influence of consumer lifestyle on brand preferences and the moderation effect of consumer characteristics.

2.1 Consumer lifestyle

Previous studies indicate that consumer lifestyle is a useful concept for marketing and advertising planning purpose. The more marketers and brand managers know and understand about their customers, the more effectively they can communicate with and serve their customers (Kucukemiroglu, 1999). Lifestyle can be defined as the patterns in which people live, spend their money and allocate their time in order to differentiate themselves from others (Kaynak & Kara, 2001; Vyncke, 2002). Consumer lifestyle therefore helps to make sense of what consumers do, why they do it, and what doing it means to them and others (Vyncke, 2002). According to Kucukemiroglu (1999) lifestyle research measures how people spend their time, what interests they have, their views of themselves and the world around them, and some basic demographic characteristics.

Before, the activities, interests and opinions (AIO) items were used for lifestyle research (Vyncke, 2002). Activities are manifests actions such as work, hobbies, social events, etc. Interest in some objects, events or topics is the degree of excitement that accompanies both special and continuing attention to it. Opinions are descriptive beliefs, for example of oneself, social issues, products, culture, etc. (Plummer, 1974). Besides AIO, the VALS2 approach is also a widely used lifestyle classification system (Lin, 2002). The

(8)

VALS2 classification considers the time and money consumers spend and links consumer demographics and purchase patterns with psychological attitudes (Lin, 2002). Although studies using the AIO or VALS2 approaches have produced insightful and meaningful data, the large battery of items required extensive surveying, analysis and resources (Vyncke, 2002). However, comparing to existing scales such as AIO and VALS2, the lifestyle scale from Green, Cordell, Betz and DiStefano (2006) uses far fewer items and therefore is better suited for a research with time constraints.

Green et al. (2006) conceptualized a practical 36-item lifestyle scale of people’s overall hobby, recreational, social, work, and special interest activities. According to Green et al. (2006) the nine different lifestyles are: Modern Life, Education & Self-learning, Watching Sports, Contributing, Home & Family, Work, Travel, Hobbies, and Nature & Environment (Table 1). Modern Life expresses the tendency to stay current with contemporary culture, health trends, up-to-date technology, and up-to-date information. Education & Self-learning indicates a desire for knowledge, skills, new experiences, and for spending time with family members. Watching Sports reflects an interest in sports, both watching and or attending sports. Contributing illustrates a propensity for charitable, civic or community involvement. Home & Family reveals an affinity for traditional activities such as cooking meals at home and spending time raising children. Work relates to business or work-oriented activities such as operating your own business and telecommuting. Travel reflects the desire to take extended vacations and to live somewhere else. Hobbies express an affinity for making, collecting or drawing things. Nature & Environment indicate an interest in nature in general (Green et al., 2006).

(9)

Table 1 Consumer lifestyle groups

Lifestyle Definition

1. Modern Life The tendency to stay current with contemporary culture, health trends, up-to-date technology, and up-to-date information

2. Education & Self-learning A desire for knowledge, skills, new experiences, and for spending time with family members

3. Watching Sports An interest in sports, both watching and/or attending sports 4. Contributing A propensity for charitable, civic or community involvement

5. Home & Family An affinity for traditional activities such as cooking meals at home and spending time raising children

6. Work Business or work-oriented activities such as operating your own business and telecommuting

7. Travel The desire to take extended vacations and to live somewhere else 8. Hobbies An affinity for making, collecting or drawing things

9. Nature & Environment An interest in nature in general

Note: The definitions are extracted from Green, Cordell, Betz and DiStefano (2006)

2.2 Brand preferences

The brand preferences construct in this study is measured using fashion brands based on Aaker’s (1997) brand personality scales (Table 2). This study will therefore focus on the brand personality construct as the focal measure of brand preferences. Brand personality can be defined as a set of human characteristics both applicable and relevant to brands (Aaker, 1997; Mulyanegara & Tsarenko, 2009). For example, Absolut Vodka personified tends to be described as a cool, hip, and contemporary 25-year old man (Aaker, 1997).

According to Park and John (2010) a brand personality can differentiate the brand from competitors. The number of brands has increased rapidly and competition has intensified, making it difficult for brands to differentiate on the basis of functional attributes alone. As a result, brands are increasingly differentiating by symbolic meanings (Siguaw, Mattila & Austin, 1999). For example, the personality traits associated with Coca Cola are cool, all-American, and real, which differentiate Coca Cola from its competitors Pepsi (being young, exciting, and hip) and Dr Pepper (being nonconforming, unique, and fun) (Pendergrast, 1993 in Aaker, 1997).

(10)

In contrast to product-related attributes brand personality tends to serve a symbolic or self-expressive function for consumers instead of a utilitarian function (Keller, 1993 in Aaker, 1997). To successfully differentiate a brand from competitors, the brand personality must be distinctive, robust, desirable and constant. Brand managers are interested in promoting a brand personality that attracts consumers’ attention such that consumers may form a preference towards the brand (Mulyanegara et al., 2009). A well-established brand personality can result in higher emotional ties to the brand, increased brand preference and usage, and consumer trust and loyalty (Siguaw et al. 1999). Sweeney and Brandon (2006) also argue that a brand’s personality can increase consumer preference for and usage of the brand, and help consumers differentiate among the innumerable products available in the marketplace.

Moreover, brand personality can be used to investigate consumers’ brand perceptions and describe brands as if they were humans (Huang et al., 2012). The symbolic use of brands is possible, because consumers can easily think about brands as if they were celebrities or famous figures and as they relate to themselves (Aaker, 1997). Moreover, Aaker and Fournier (1995) argue that a brand can function as a character, partner and person. If brands are one of the resources used to reflect consumers’ identity, consumers will choose brands that reflect their personalities (Huang et al., 2012). According to Park and John (2010) consumers are attracted to brands with distinctive and appealing personalities when they wish to express, affirm or enhance their sense of self.

Brand personality is based on traits from The Big Five model that could be attributed to brands (Mulyagenara et al., 2009). The five different brand personality dimensions are Sincerity (e.g. honest, friendly), Excitement (e.g., trendy, cool), Competence (e.g., intelligent, hardworking), Sophistication (e.g., glamorous, feminine) and Ruggedness (e.g., outdoorsy, tough) (Aaker, 1997; Park & John, 2010). Although both the brand personality scale and the consumer personality scale yield a five-factor structure, only the three factors Sincerity,

(11)

Excitement and Competence appear in both personality structures (Huang et al., 2012). Sincerity and Agreeableness both capture the idea of warmth and acceptance. Excitement and Extroversion both connote the notions of sociability, energy, and activity. Competence and Conscientiousness both include responsibility, dependability, and security (Aaker, 1997). The dimensions Sophistication and Ruggedness differ from any of the Big Five, since they are not part of human personality, but something that individuals desire and not necessarily have (Aaker, 1997). Human and brand personalities differ in terms of how they are formed (Aaker, 1997). Human personality is formed by behavior, physical characteristics, attitudes and beliefs, and demographic characteristics, while brand personality is formed and influenced by any direct or indirect contact between the consumer and the brand (Plummer, 1985 in Aaker, 1997). Personality traits come to be associated with a brand in a direct way by the people associated with the brand such as the company’s employees, CEO, and the brand’s endorsers. Moreover, personality traits come to be associated with a brand in an indirect way through product-related attributes, product category associations, brand name, symbol or logo, advertising style, price, and distribution channel (Batra, Lehmann, and Singh, 1993 in Aaker, 1997).

(12)

Table 2 Brand personality dimensions and related fashion brands

Brand personality Characteristics Fashion brands

Sincerity Down-to-earth: family oriented, small town Honest: sincere, real

Wholesome: original

Cheerful: friendly, sentimental

Esprit Gap

H&M Conscious Excitement Daring: trendy, exciting

Spirited: cool, young Imaginative: unique

Up-to-date: independent, contemporary

River Island Topshop / Topman Abercrombie & Fitch Competence Reliable: hard working, secure

Intelligent: technical, corporate Successful: leader, confident

Calvin Klein Hugo Boss Filippa K Sophistication Upper-class: good-looking, glamorous

Charming: feminine, smooth

Dior Chanel Prada Ruggedness Outdoorsy: masculine, western

Tough: rugged

The North Face Timberland Levi’s

2.3 Consumer characteristics

The consumer characteristics construct in this study is measured using the human personality scales. Human personality is formed by behavior, physical characteristics, attitudes and beliefs, and demographic characteristics (Plummer, 1985 in Aaker, 1997). This study will focus on the Big Five construct as the focal measure of consumer characteristics. Many researchers have attempted to develop a universal human personality framework to explain individual differences, but the Big Five model of Costa and McCrae (1992) is the most widely used and extensively researched model of human personality (Mulyanegara et al., 2009; Gosling, Rentfrow & Swan Jr., 2003). Furthermore, the dimensionality of the Big Five has been found to generalize across virtually all cultures and remains fairly stable over time (McCrae & Costa, 1997; Judge, Higgings, Thoresen & Barrick, 1999). However, Costa and McCrae’s 240-item NEO Personality Inventory takes about 45 minutes to complete, and thus is too lengthy for this research purpose (Gosling et al., 2003). Therefore, this study will use a

(13)

brief measure of the Big-Five personality domains, namely the Ten-Item Personality Inventory (TIPI) by Gosling et al. (2003).

Personality can be seen as a configuration of an individual’s cognition, emotion and motivation, which activate consumer behavior and reflects how the consumer adjusts to the environment and life experiences (Triandis & Suh, 2002; Huang et al., 2012). According to Mulyanegara et al. (2009) personality can be defined as “the intrinsic organization of an individual’s mental world that is stable over time and consistent over situations”. The Big Five model describes a consumer’s personality by cognitive, emotional and behavioral elements that are distinctive in their particular combination (Mulyanegara & Tsarenko, 2009). The different Big Five dimensions are: Neuroticism, Extroversion, Openness to Experience, Agreeableness, and Conscientiousness (Table 3). Furthermore, all the Big Five dimensions are bipolar (e.g., Extraversion vs. Introversion). Consumers can express all five dimensions, however they may score quite highly on one or several dimensions and lower on others (McCrae & Costa, 1990 in Mulyanegara et al., 2009). As noted before, only the three factors Agreeableness, Extroversion, and Conscientiousness appear in both the human personality and brand personality structures (Huang et al., 2012).

Neuroticism is the most pervasive trait across personality measures and is prominent in nearly every measure of human personality (Costa and McCrae, 1988 in Judge et al., 1999). Neuroticism refers to a lack of positive psychological adjustment and emotional stability, so individuals with this personality are more likely to experience negative moods and physical symptoms. Like Neuroticism, the dimension Extraversion also appears in most human personality measures (Judge et al., 1999). Extraverts are socially oriented, but also dominant and active (Judge et al., 1999). Extraverts therefore often have a greater number of close friends and are more likely to take on leadership roles (Watson & Clark, 1997 in Judge et al., 1999). Openness to experience is characterized by intellectance and unconventionality (Judge

(14)

et al., 1999). Agreeable individuals are cooperative and likeable (Judge et al., 1999). Conscientiousness is related to an individual’s degree of self-control, need for achievement, order and persistence (Costa, McCrae & Dye, 1991 in Judge et al., 1999).

The brief measure of the Big Five personality domains by Gosling et al. (2003) used in this study, breaks each of the five different personality dimensions into bipolar pairs of traits: Extraversion (extraverted and enthusiastic vs. reserved and quiet), Agreeableness (sympathetic and warm vs. critical and quarrelsome), Conscientiousness (dependable and self-disciplined vs. disorganized and careless), Neuroticism (anxious and easily upset vs. emotionally stable and calm), and Openness to Experience (openness to new experiences and complex vs. conventional and uncreative).

Table 3 The Big Five human personality dimensions

The Big Five Dimensions Definition Corresponding Brand personality

Neuroticism A tendency to be emotionally unstable Extraversion A tendency to seek for stimulation in life and

enjoy activities in groups

Excitement Openness to Experience A tendency to be open-mined to new ideas

Agreeableness An interpersonal tendency that describes altruisticity

Sincerity Conscientiousness A tendency to be self-disciplined Competence

Note: The definitions are extracted from Costa & McCrae (1992) in Huang et al. (2012: 335)

2.4 Conclusion

Previous studies about lifestyle suggest that a consumer’s lifestyle influences the consumption pattern, because lifestyle helps to make sense of how consumers spend their money, allocate their time, and what interests they have (Kucukemiroglu, 1999; Vyncke, 2002). Besides consumer lifestyle, other factors like consumer characteristics could also influence consumption patterns or brand preferences. The Big Five model seems to affect brand preferences, since consumers use brands to express their personality and therefore prefer

(15)

brands that are similar to their own personalities (Mulyanegara et al., 2009; Huang et al., 2012).

However, former studies have not investigated the relationship between consumer lifestyle and preferences towards fashion brands thoroughly before. Therefore, this study aims to contribute to the lifestyle literature, the branding literature, the consumer behavior literature, and to fashion research. Furthermore, it can help brand managers to communicate their fashion brand in an effective way to attract consumers with a particular lifestyle or with specific consumer characteristics. Not only can the right media be identified to reach specific consumers, but similar lifestyle themes may also be included in the message and executional cues (Orth, McDaniel, Shellhammer & Lopetcharat, 2004).

The current study tries to fill this gap in the literature by attempting to find an answer to the main research question: How does consumer lifestyle influence preferences towards

fashion brands and how is this relationship influenced by consumer characteristics? To

provide an answer on this research question the important determinants consumer lifestyle, fashion brand preferences and consumer characteristics will be explored.

(16)

3. Conceptual model

The conceptual model (Figure 1) is based on the main determinants consumer lifestyle, consumer characteristics, and their influence on fashion brand preferences. This conceptual model will be applied to well-known fashion brands in the Netherlands that have different brand personalities. Brand managers in the fashion industry can use the conceptual model to find effective ways to target and attract consumers with particular lifestyles and consumer characteristics based on brand personality.

Figure 1 Conceptual model

3.1 The influence of consumer lifestyle on fashion brand preferences

As stated before lifestyle is defined as the patterns in which people live, spend their money and allocate their time in order to differentiate themselves from others (Kaynak and Kara, 2001; Vyncke, 2002). Since consumers use brands to express their identities, many companies try to focus on how their products fit into a consumer’s lifestyle instead of focusing on functional attributes (Chernev, Hamilton & Gal, 2011). Helman and Chernatony (1999) argue that a brand provides a set of added values that have symbolic value and meaning for the lifestyle of a particular consumer group. Brands can be used to communicate membership in

(17)

particular social or professional groups, through both the use of desirable brands and the avoidance of undesirable brands (Chernev et al., 2011).

According to Chernev et al. (2011) consumer brand preferences are a function of the activities consumers were involved in prior to evaluating a given brand and the degree to which these prior activities afforded the opportunity to express their identities. Moreover, consumers’ trust on a given brand to express their identity not only depends on brands in the same category, but is also a function of brands in unrelated categories and even non-brand means of self-expression such as hobbies (Chernev et al., 2011). Thus, a consumer’s preference for a particular brand is likely to be a function of the strength of the consumer’s associations with cross-category self-expressive brands or even whether the consumer has recently engaged in self-expressive activities such as creating art, updating a Facebook profile, and donating to charity (Chernev et al., 2011). Therefore, this study predicts that consumer lifestyle is correlated with fashion brand preferences, which indicates that consumer lifestyle will influence the consumers’ preference towards fashion brands.

Hypothesis 1: Consumer lifestyle will influence consumers’ preferences towards specific fashion brands.

3.2 The moderation effect of consumer characteristics

As stated before, personality can be defined as “the intrinsic organization of an individual’s mental world that is stable over time and consistent over situations” (Mulyanegara et al., 2009). Many researchers have argued that consumers use brands to express their personality and therefore prefer to use brands that are similar to their own personalities (Mulyanegara et al., 2009; Huang et al., 2012). This can be explained by the fact that consumers like to express the type of person they are or the type of person they think they are through the use of brands (Mulyanegara & Tsarenko, 2009; Huang et al., 2012). According to Park and John (2010)

(18)

brand personality is appealing to consumers who want to express, affirm or enhance their sense of self. They argue that consumers with certain implicit beliefs about their personalities (entity theorist) are affected by brand experiences, resulting in more positive perceptions of themselves on personality traits associated with the brands they use. Furthermore, when consumers find consistency between their personality and the brand characteristics, they begin developing a relationship with the brand as if they were interacting with an actual person (Aaker & Fournier, 1995; Mulyanegara & Tsarenko, 2009). Moreover, recent studies concluded that in order for a brand to be successful it must generate perceived attributes that are consistent with the personality of the target consumers (Mulyanegara & Tsarenko, 2009).

However, only the human personality dimensions Agreeableness, Extroversion and Conscientiousness were found to correlate directly with the brand personality dimensions Sincerity, Excitement and Competence. Both Agreeableness and Sincerity capture the idea of acceptance and warmth (Aaker, 1997). Extroverts prefer Excitement brands to reflect their outgoing characteristics and consumers who are Conscientious prefer Competence brands to reflect their reliable characteristics (Aaker, 1997; Mulyanegara & Tsarenko, 2009). Therefore, it is expected that consumers who are dominant on a particular dimension of the Big Five will prefer a brand personality which reflects that dimension or is close to it (Mulyanegara & Tsarenko, 2009). According to Aaker (1997) human and brand personalities differ in terms of how they are formed. Human personality is formed by behavior, physical characteristics, demographic characteristics, attitudes and beliefs, while brand personality is formed by any direct or indirect contact between the consumer and the brand (Plummer, 1985 in Aaker, 1997).

Moreover, Huang et al. (2012) argue that consumers reflect their personalities by the brands they use, but the relationship between brand preferences and the symbolic dimensions Extraversion, Agreeableness, Neuroticism and Openness to Experience is much stronger than

(19)

the relationship with a functional dimension like Conscientiousness. The pattern of this relationship remains consistent across various products, which implies that consumers choose brands with similar characteristics to theirs across both symbolic and utilitarian products (Huang et al., 2012). This study agrees with Aaker & Fournier (1995), Mulyanegara et al., (2009), Mulyanegara and Tsarenko (2009), and Huang et al. (2012), but is more extensive and predicts that there will be an interaction between consumer lifestyle and consumer characteristics on fashion brand preferences, which indicates that consumer characteristics will influence the fashion brand preference of consumers with a particular lifestyle.

Hypothesis 2: Consumer characteristics will moderate the relationship between consumer lifestyle and fashion brand preferences.

(20)

4. Methodology

This section will first discuss the research design. Thereafter the sample used in the survey and the data will be discussed, followed by the measurements of the variables consumer lifestyle, fashion brand preferences, and consumer characteristics. Finally, the data analyses will be discussed.

4.1 Research design

For this study an anonymous self-administered questionnaire about the variables consumer lifestyle, fashion brand preferences and consumer characteristics is used in order to collect the data needed to test the hypothesis. The questionnaire comprised four parts. The first part contains a number of questions about the consumers’ demographics, such as gender, age, nationality, income and education. The second part included the lifestyle scale from Green, Gordell, Betz & DiStefano (2006). The third part was about preferences towards fashion brands based on Aaker’s (1997) brand personality dimensions. The last part comprised a brief Big Five personality scale by Gosling, Rentfrow & Swann (2003).

Survey questionnaires are appropriate for this study, because it allows the collection of quantitative data and it can be used to suggest possible reasons for particular relationships between variables (Saunders, Lewis & Thornhill, 2009, p.144). The survey strategy also allows the collection of a large amount of data from a sizeable population in a quick and affordable way (Saunders et al., 2009, p.144). Furthermore, a survey makes it easy to compare the answers of a large amount of people, because when using a survey questionnaire these data are standardized (Saunders et al., 2009, p.144).

A self-administered questionnaire is used for this study, because self-administered questionnaires increase the chance of anonymity, which in turn reduces the participant or subject bias and improves the reliability of the data (Saunders et al., 2009, p.156).

(21)

Furthermore, a self-administered questionnaire with increased anonymity also reduces socially responsible answers (Saunders et al., 2009, p.365). Moreover, self-administered questionnaires give respondents the opportunity to answer the questions at the time that is most convenient to them, which will increase the willingness to participate.

However, a survey strategy also has some methodological limitations. One limitation is the fact that there is only one chance to collect the data, because it is often difficult to identify respondents or to return to collect additional information (Saunders et al., 2009, p.366). Therefore, designing the survey questionnaire must be done very carefully prior to data collection. Another limitation is that, due to time constraints, this study is cross-sectional with the focus on a specific phenomenon at a specific time. Therefore, it will not be possible to draw conclusions about causality based on the findings of a survey (Saunders et al., 2009, p.155).

4.2 Sample

This study focuses on Dutch and international consumers with different lifestyles. To select the cases from this population a non-probability method is used, where the probability of each case being selected from the total population is unknown. Despite the fact that a non-probability method is used it is still possible to generalize about the population, however, not on statistical grounds (Saunders et al., 2009, p.213). The method used in this study is self-selection sampling, since each case is allowed to identify their desire to take part in the online survey (Saunders et al., 2009, p.217). However, the fact that individuals can choose whether they want to participate or ignore the online survey invitation can lead to a systematic bias.

To increase the generalizability of the data results, this study aimed for a sample size as large as possible (Saunders et al., 2009, p.217). Moreover, this study aimed for a sample with a wide range of lifestyles, varied from ‘Home & Family’ to ‘Work’ and from ‘Modern

(22)

Life’ to ‘Travel’ (Green et al., 2006). The data results would be more reliable and generalizable when the sample entails consumers of different backgrounds and ages. However, because of the lack of time and money, it was not possible to draw large samples as used in previous studies about this subject. The timeframe of the data collection period for this study was three weeks. In order to make sure the sample is normally distributed and generalizable, a sample size of at least 100-150 respondents was desirable.

4.3 Data collection

The respondents for the Internet survey questionnaire were recruited in several ways. Friends and acquaintances were approached by e-mail or through the social media Facebook and asked to fill out and also to distribute the Internet survey. The e-mail invitation and Facebook message included a direct web link for the self-administered questionnaire and a short and clear description, which explains the purpose of the Internet survey from the University of Amsterdam and the confidential and voluntary nature of participation. Furthermore, respondents were recruited through the professional network site LinkedIn, where it is also possible to share the link to the survey with a short description in different groups, such as ‘Economics and Business Network’ and ‘Universiteit van Amsterdam’. Moreover, some respondents were asked to fill out the survey on an iPad while traveling by train or waiting to board at Amsterdam Schiphol.

The questionnaire is written in English, because this study focuses on both Dutch and international consumers. To administer the questionnaire an Internet-mediated method is used, since online surveys can reach a lot of respondents in a relatively short time (Wright, 2005). Besides that, online surveys may also save time by allowing researchers to collect data while working on other tasks (Wright, 2005). Furthermore, an online survey is an easy and appropriate way to contact groups and individuals with different lifestyles. To develop the

(23)

self-administered questionnaire the tool Qualtrics is used. This Internet-mediated method also saves a lot of time, since the data does not have to be entered manually (Saunders et al., 2009, p.365). Moreover, this Internet-mediated method saves money as well, since there are no printing costs as with pen and paper surveys (Wright, 2005). Besides advantages the Internet-mediated method also has a disadvantage. Collecting data through the Internet can lead to a systematic bias, since individuals engage in self-selection and choose whether they want to participate or ignore the online survey invitation (Wright, 2005).

4.4 Measurements

In the survey the consumer lifestyle, consumer characteristics, and fashion brand preferences items were administered in English (Appendix), since the focus in this study is on both Dutch and international consumers. In the survey behavioral variables are used to measure consumer lifestyle, since behavioral variables contain data on what people did in the past or do now (Saunders et al., 2009, p.368). Opinion variables are used to measure the brand preferences, since these variables record how respondents feel about something (Saunders et al., 2009, p.368). Moreover, attribute variables are used to measure the consumer characteristics and consumer demographics, since attribute variables contain data about the respondents’ characteristics that the respondents possess rather than things they do (Saunders et al., 2009, p.368). Except from the demographics and the variable fashion brand preferences, all the variables were measured with borrowed measurements.

This study used rating questions to measure the variables consumer lifestyle, fashion brand preferences and consumer characteristics, because rating questions are appropriate to collect behavioral, opinion, and attribute data (Saunders et al., 2009, p.378). Responses for the consumer lifestyle, fashion brand preferences, and consumer characteristics items were given on a 1-7 likert scale (1=’strongly disagree’ to 7=’strongly agree’). These questions contain an

(24)

odd number of points on the rating scale, which allows the respondent to ‘sit on the fence’ by ticking the middle category (Saunders et al., 2009, p.379). Besides rating questions, this study also used category questions and open questions to measure the demographics as control variables. Category questions are questions where only one response can be selected from a given set of categories (Saunders et al., 2009, p.376).

The independent variable consumer lifestyle was measured with the recreational and environment’s lifestyle scale by Green, Gordell, Betz & DiStefano (2006). Green et al. (2006) identified 36 items that capture people’s major activities which are grouped under the nine factors: Modern Life, Education & Self-learning, Watching Sports, Contributing, Home & Family, Work, Travel, Hobbies, and Nature & Environment. The respondents who rated these items had to answer questions like: ‘I read news, business, or professional magazines’ and ‘I take vacations away from home at least once a year’. The dependent variable fashion brand preferences was measured using fifteen fashion brands based on Aaker’s (1997) five brand personalities Sincerity, Excitement, Competence, Sophistication, and Ruggedness. The respondents who rated these brand preferences items had to answer questions like: ‘I prefer the fashion brand Levi’s’, ‘I prefer the fashion brand River Island’ and ‘I prefer the fashion brand Hugo Boss’. In order to find fashion brands that match with the brand personalities, a pre-test was held among 20 respondents. The respondents were asked to assign 25 different fashion brands to one particular brand personality. For each brand personality the three fashion brands with the highest score were selected to measure the dependent variable brand preferences in the final survey.

Due to time constraints the moderator variable consumer characteristics was measured using a very brief measure of the Big Five personality domains (Extroversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to Experience), namely the Ten-Item Personality Inventory (TIPI) by Gosling, Rentfrow & Swann (2003). The

(25)

respondents who rated these ten items had to answer questions like: ‘I see myself as extraverted, enthusiastic’ and ‘I see myself as reserved, quiet’. Besides the moderator variable consumer characteristics other factors may also affect the relationship between consumer lifestyle and fashion brand preferences. Therefore, the demographic variables age, gender, education, and income are used as control variables. The control variables were measured with category questions and an open question developed by myself. The respondents had to answer the open question: ‘What is your age?’ and category questions like: ‘What is your gender? 1) male, 2) female’ and ‘What is the highest level of education you have completed?’ 1) No education completed, 2) Primary School, 3) High School, 4) Intermediate Vocational Education, 5) University of Applied Science, 6) Bachelor’s Degree, 7) Master’s Degree, 8) PhD’.

4.5 Analyses

To test the hypotheses the collected data are analyzed using statistical program SPSS. First all the variables were checked for missing data and counter-indicative items. Counter-indicative items had to be recoded in order to be able to conduct the reliability analysis. Then, the reliability analysis was conducted to test the internal consistency and to see if it was necessary to delete items for the variables ‘consumer lifestyle’, ‘fashion brand preferences’ or ‘consumer characteristics’. Subsequently, the descriptive statistics (the mean and standard deviations) and correlations of the variables ‘consumer lifestyle’, ‘fashion brand preferences’, ‘consumer characteristics’ and control variables were calculated. Thereafter the regression test was conducted to see if the independent variables ‘consumer lifestyle’ and ‘consumer characteristics’ predicts the dependent variable ‘fashion brand preferences’. Then, it was tested if the control variables age, gender, education, and income also had an effect on the dependent variable ‘fashion brand preferences’ and the independent variables. Furthermore, it

(26)

was tested if the variable ‘consumer characteristics’ moderates the relationship between the variables ‘consumer lifestyle’ and ‘fashion brand preferences’. This moderation effect was tested by means of interaction effects between ‘consumer lifestyle’ and ‘consumer characteristics’ on ‘fashion brand preferences’. To compute the interaction variable the variable ‘consumer lifestyle, ‘consumer characteristics’ and ‘fashion brand preferences’ had to be centered first.

(27)

5. Results

The results of the research discussed in the section methodology will be provided in this section. First the respondents from the survey will be described. Thereafter, the reliability of scales is shown by means of the Cronbach’s Alpha for the consumer lifestyle, consumer characteristics, and fashion brand preferences items. Then, some descriptive statistics and correlations will be discussed. Finally, the results from the regression analysis will be presented.

5.1 Participants

A total of 181 respondents agreed to participate in this study. However, from the 181 respondents only 152 respondents completed the survey questionnaire and 29 respondents did not complete the survey questionnaire. From these 29 respondents all the consumer lifestyle items, fashion brand preferences items, and/or the consumer characteristics items were missing. Therefore, these 29 respondents had to be excluded from the analyses.

Even after excluding the 29 respondents there were still some items missing. From respondents 7, 9, 13, 18, 20, 26, 27, 45, 70, 78, 79, 87, 91, 98, 105, 108, 110, 114, 130, and 132 at least one out of 36 consumer lifestyle items was missing. From respondents 7, 16, 24, 33, 46, 68, 79, 96, 99, 101, 138, and 144 one out of fifteen fashion brand preferences items was missing. Furthermore, from respondents 26, 51, and 58 one out of ten consumer characteristics items was missing. However, because of the little number of total respondents these respondents were not excluded from the analyses. When doing the analyses in SPSS the ‘exclude cases pairwise’ box was ticked for the missing items of those respondents. When ticking the ‘exclude cases pairwise’ box, only the cases without missing data in each pair of variables will be analyzed.

(28)

The final dataset therefore included 152 respondents (N = 152). The respondents were between the ages 13 and 58 with an average of 26.17 years. However, only 114 out of 152 respondents filled out their age and 38 participants forgot or did not want to fill out their age. From the 152 respondents 90 respondents were female (59%) and 62 respondents were male (41%). Regarding education most respondents have completed their Master’s Degree (41%). The remaining respondents completed their Bachelor’s Degree (23%), High School (16%), University of Applied Science (11%), Intermediate Vocational Education (7%), PhD (1%), and Primary School (1%). Regarding nationality most respondents were from the Netherlands (81%). The remaining respondents were from Germany (3%), Czech Republic (3%), Greece (2%), Vietnam (2%), but also from Belgium, United Kingdom, Finland, France, Italy, Romania, Slovakia, Australia, New Zealand, Indonesia, and South-Korea (1%). Regarding income most respondents had a gross monthly income of less than €1000,- (51%). The remaining respondents had a gross monthly income between €2000-€2999 (18%), between €1000-€1999 (18%), more than €4000,- (8%), and between €3000-€3999 (5%).

5.2 Reliability

To test whether the items for the variables consumer lifestyle, fashion brand preferences, and consumer characteristics were reliable, the reliability analysis was conducted. Reliability refers to the extent to which your data collection techniques will yield consistent findings, or in other words, the extent to which the measures will yield the same results in other occasions (Saunders et al., 2009, p.156). However, it was not possible to conduct the reliability analysis for the control variables, since these were measured with only one item. Furthermore, some counter-indicative consumer lifestyle items and consumer characteristics items had to be recoded in order to be able to conduct the reliability analysis. The Cronbach’s Alphas of the

(29)

variables consumer lifestyle, fashion brand preferences, and consumer characteristics are presented in Table 3.

The Cronbach’s Alphas for the variables consumer lifestyle (α = 0.772) and brand preferences (α = 0.848) were both highly reliable. The Cronbach’s Alpha of both variables could not significantly increase when one of the items would be deleted, but in this case a higher Cronbach’s Alpha is not necessary. However, the Cronbach’s Alpha for consumer characteristics was unacceptable, since the Cronbach’s Alpha was <0.5 (α = 0.201). The variable consumer characteristics could significantly increase (α = 0.299) if the item ‘I see myself as critical, quarrelsome’ would be deleted, however the Cronbach’s Alpha would still not be acceptable.

The reliability analysis was also conducted for the items of the nine separate consumer lifestyles. The Cronbach’s Alpha for the lifestyle Modern Life was poor (α = 0.509) and could significantly increase (α = 0.563) when the item ‘I visit a fitness club or otherwise exercise at least 3 times a weak’ would be deleted. However, the new Cronbach’s Alpha for the lifestyle Modern Life would still be poor. The Cronbach’s Alpha for Education & Self-learning was also unacceptable (α = 0.241) and could not significantly increase when one of the items would be deleted. The Cronbach’s Alpha for the lifestyle Watching Sports was found to be highly reliable (α = 0.734). The Cronbach’s Alpha for the lifestyle Contributing was found to be poor (α = 0.538). The Cronbach’s Alpha for the lifestyle Home & Family was unacceptable (α = 0.452) and could not significantly increase. The Cronbach’s Alpha for the lifestyle Work was acceptable (α = 0.691). The Cronbach’s Alpha for the lifestyle Travel was also unacceptable (α = 0.433), but could not significantly increase. The Cronbach’s Alpha for the lifestyle Hobbies was acceptable (α = 0.670). The Cronbach’s Alpha for Nature & Environment was unacceptable (α = 0.364), but could become more acceptable (α = 0.593) when the item ‘I eat out in restaurants, including fast food, or order take-out food at least 2

(30)

times a week’ would be deleted. Therefore, the Nature & Environment lifestyle item ‘I eat out in restaurants, including fast food, or order take-out food at least 2 times a week’ is deleted for further analysis.

Furthermore, the reliability analysis was conducted for the fashion brand preferences items based on the five different brand personalities. The Cronbach’s Alpha for preferences towards Sincere brands was unacceptable (α = 0.465) but could become more acceptable (α = 0.587) when the item ‘I prefer the fashion brand H&M Conscious’ would be deleted. The Cronbach’s Alpha for preferences towards Exciting brands was acceptable (α = 0.662) and could become highly reliable (α = 0.719) when the item ‘I prefer the fashion brand Abercrombie & Fitch’ would be deleted. The Cronbach’s Alpha for preferences towards Competent brands was acceptable (α = 0.632) and could become highly reliable (α = 0.705) when the item ‘I prefer the fashion brand Filippa K’ would be deleted. The Cronbach’s Alpha for preferences towards Sophisticated brands was highly reliable (α = 0.888). The Cronbach’s Alpha for preferences towards Rugged brands was poor (α = 0.594) but could not significantly increase when one item would be deleted. Thus, the items ‘I prefer the fashion brand H&M Conscious’, ‘I prefer the fashion brand Abercrombie & Fitch’ and ‘I prefer the fashion brand Filippa K’ are deleted in order to increase the internal consistency.

Finally, the reliability analysis was also conducted for the five different consumer characteristics groups. The Cronbach’s Alpha for Extraversion was found to be highly reliable

(α = 0.741). The Cronbach’s Alpha for Agreeableness was found to be unacceptable (α = 0.360), but could not significantly increase. The Cronbach’s Alpha for Conscientiousness

was also found to be unacceptable (α = 0.178). The Cronbach’s Alpha for Neuroticism was found to be acceptable (α = 0.635). The Cronbach’s Alpha for Openness to Experience was found to be unacceptable (α = 0.316), but could not significantly increase.

(31)

5.3 Descriptive statistics and correlations

The correlation analysis was conducted to quantify the intensity and meaning of the relationship between the main variables consumer lifestyle, consumer characteristics, and fashion brand preferences (Table 4). As predicted earlier, the independent variable consumer lifestyle is a moderate predictor of the dependent variable brand preferences (r = .35, p < .01). This means that the consumers’ lifestyle does influence the preferences of consumers towards specific fashion brands. Moreover, there is a moderate correlation between consumer lifestyle and the moderator variable consumer characteristics (r = .30, p < .01). This means that a consumers’ lifestyle has an effect on the personality traits that consumers possess. Furthermore, the moderator variable consumer characteristics is weakly correlated with the dependent variable brand preferences (r = .23, p < .01). This indicates that the consumers’ personality traits have an influence on their preferences towards specific fashion brands.

As shown in Table 4, the demographics such as age, gender, education, and income are not significant predictors of the dependent variable brand preferences. For the first control variable age no correlations were found. For the second control variable gender a weak correlation was found with the moderator variable consumer characteristics (r = .20, p < .05). This means that gender influences the personality traits that the consumer possesses. In this case female consumers appear to possess more different personality traits than male consumers do. For the third control variable education a very weak correlation was found with the independent variable consumer lifestyle (r = .17, p < .05). This indicates that consumers with higher education possess more different lifestyles than less educated consumers do. For the fourth and last control variable income a very weak correlation was found with the independent variable consumer lifestyle (r = .18, p < .05). This means that consumers with a higher income possess more different lifestyles than consumers with a lower income do.

(32)

Table 4 Means, Standard Deviations, and Correlations Variables M SD 1 2 3 4 5 6 7 1.Age 2.15 0.79 - 2.Gender 1.59 0.49 .18* - 3.Education 5.67 1.53 .22* -.04 - 4.Income 1.99 1.26 .45** -.12 .32** - 5.Consumer Lifestyle 3.86 0.60 .09 -.04 .17* .18* (.77) 6.Consumer Characteristics 4.52 0.50 .11 .20* -.07 .07 .30** (.85) 7.Brand Preferences 3.64 1.02 -.08 .16 -.04 -.07 .35** .23** (.20)

**Correlation is significant at the 0.01 level (2-tailed) *Correlation is significant at the 0.05 level (2-tailed)

In Table 5 the means, standard deviations, and correlations of the demographics, different lifestyle groups, different consumer characteristics groups, and different fashion brand preferences groups are shown. Based on the results in Table 4 it can be assumed that the consumers’ lifestyle has an effect on the preferences of consumers towards specific fashion brands (r = .35, p < .01). The results in Table 5 show that consumers with a Modern Life lifestyle have a weak preference for fashion brands that are Sincere (r = .27, p < .01), Competent (r = .25, p < .01), and Rugged (r = .23, p < .01). Consumers with an Education & Self-learning lifestyle have a weak preference for Exciting (r = .26, p < .01) and Sincere

(r = .20, p < .05) fashion brands, and seem to have a negligible preference for Sophisticated (r = .18, p < .05) and Rugged (r = .17, p < .05) fashion brands. Consumers with the lifestyle

Watching Sports have a weak preference for fashion brands that are Sincere (r = .23, p < .01), Competent (r = .21, p < .01), and Rugged (r = .20, p < .05), but appear to dislike Sophisticated fashion brands (r = -.17, p < .05). Consumers with a Contributing lifestyle tend to prefer Sincere fashion brands (r = .17, p < .05). Consumers with a Home & Family lifestyle prefer fashion brands that are Sophisticated (r = .25, p < .01) while consumers with the lifestyle Work prefer Competent brands (r = .21, p < .01). Consumers with a Travel lifestyle have a moderate preference for fashion brands that are Competent (r = .31, p < .01), a weak

(33)

preference for brands that are Rugged (r = .27, p < .01), Sincere (r = .24, p < .01), and

Exciting (r = .21, p < .05), and a negligible preference for Sophisticated brands (r = .19, p < .05). Consumers with the lifestyle Hobbies appear to prefer fashion brands that

are Rugged (r = .20, p < .05). Finally, consumers with a Nature & Environment lifestyle have a weak preference for Sincere fashion brands (r = .28, p < .01).

Furthermore, the results in Table 4 suggest that the moderator variable consumer

characteristics is correlated with the independent variable consumer lifestyle (r = .30, p < .01). In Table 5 a weak correlation was found between the consumer

characteristic Agreeableness and the lifestyles Contributing (r = .26, p < .01) and Home & Family (r = .21, p < .01) and a negligible correlation between Agreeableness and the lifestyle Education & Self-learning (r = .18, p < .05). This indicates that Agreeable consumers often have a Contributing or Home & Family lifestyle. A moderate correlation was found between

the consumer characteristic Conscientiousness and the lifestyle Home & Family (r = .30, p < .01), which means that Conscientious consumers often have a Home & Family

lifestyle. A moderate correlation was found between the consumer characteristic Openness to Experience and the lifestyle Travel (r = .33, p < .01) and weak correlations between Openness

to Experience and the lifestyles Hobbies (r = .29, p < .01), Nature & Environment (r = .26, p < .01), Modern Life (r = .25, p < .01), Education & Self-learning (r = .20, p < .05),

and Work (r = .20, p < .05). This indicates that consumers who are Open to Experiences often have a Travel lifestyle, but also seem to have the lifestyles Hobbies, Nature & Environment, Modern Life, Education & Self-learning, and Work. However, there were no correlations found between consumer lifestyle and the consumer characteristics Extraversion or Neuroticism, which means that consumers with these characteristics do not necessarily have a specific lifestyle.

(34)

Based on the results in Table 4 it can also be concluded that the moderator variable consumer characteristics (Big Five) seems to influence consumers’ fashion brand preferences (r = .23, p < .01). Table 5 shows that Extravert consumers appear to prefer Competent fashion brands (r = .19, p < .05). Agreeable consumers (r = .18, p < .05) and Conscientious consumers (r = .27, p < .01) both prefer Sophisticated fashion brands. However, Neurotic consumers do not seem to prefer fashion brands with a specific brand personality. Finally, consumers with the characteristic Openness to Experience appear to prefer Rugged fashion brands (r = .19, p < .05) and dislike Sincere fashion brands (r = -.18, p < .05).

As regards to the demographics, a very weak correlation was found between the first control variable age and the lifestyle Home & Family (r = .19, p < .05). This indicates that consumers with a Home & Family lifestyle are often of an older age. Furthermore, consumers with the characteristic Agreeableness are also often of an older age (r = .21, p < .05). However, a moderate negative correlation was found between age and preferences towards Exciting brands (r = -.30, p < .01). This indicates that the older the consumer, the greater the chance that they dislike Exciting brands.

A negative moderate correlation was found between the second control variable gender and the lifestyle Watching Sports (r = -.36, p < .01) and a very weak negative correlation between gender and Work (r = -.19, p < .05). This means that consumers with the Watching Sports or Work lifestyles are often men. Furthermore, gender weakly correlates with the lifestyle Home & Family (r = .21, p < .05), which indicates that consumers with this

lifestyle are often women. Moreover, Conscientious (r = .31, p < .01) and Agreeable (r = .28, p < .01) consumers also often appear to be women. Besides that, gender influences

consumers’ preferences towards fashion brands. Sophisticated (r = .33, p < .01) and Exciting (r = .26, p < .01) fashion brands are often preferred by women, while Competent brands (r = -.24, p < .01) are often preferred by men.

(35)

A weak correlation was found between the third control variable education and the lifestyle Modern life (r = .26, p < .01), which indicates that consumers with this lifestyle often have completed a higher level of education. Furthermore, a weak negative correlation was found between education and the consumer characteristic Neuroticism (r = -.21, p < .01), which means that consumers who are Neurotic appear to be less educated. However, no correlation was found between education and consumers’ preferences towards specific fashion brands.

The fourth and last control variable income has a moderate correlation with the lifestyles Work (r = .31, p < .01) and a very weak correlation with the lifestyle Modern Life (r = .19, p < .05), which means that consumers with the lifestyles Work or Modern Life often seem to have a higher income. Moreover, a very weak negative correlation was found between income and the consumer characteristic Neuroticism (r = -.19, p < .05), which indicates that Neurotic consumers appear to have a lower income. Regarding fashion brand preferences, it can be said that the higher the consumers’ income, the greater the chance that they dislike Exciting brands (r = -.16, p < .05).

(36)

32 ! b le 5 Me an s, S ta nd ar d D ev ia ti on s, a nd C or re la ti on s es M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 2. 15 0. 79 - r 1. 59 0. 49 .1 3 - at ion 5. 67 1. 53 .2 1* -.0 4 - e 1. 99 1. 26 .5 4* * -.1 2 .3 2* * - rn L if e 4. 74 0. 74 .1 4 -.0 3 .2 6* * .1 9* (. 51 ) at ion & ar nin g 5. 21 0. 81 -.0 8 .1 3 .1 4 -.1 3 .4 5* * (. 24 ) chi ng 3. 94 1. 87 -.1 3 - .36* * .0 2 -.0 7 .2 5* .1 6* (. 73 ) ri bu ti ng 2. 58 1. 27 .0 5 .0 6 .1 0 .1 0 .3 0* .0 4 .1 6 (. 54 ) e & y 3. 78 1. 07 .1 9* .2 1* .0 2 .1 3 .3 6* * .0 7 .1 2 .2 7* * (. 45 ) or k 2. 57 1. 58 .1 1 -.1 9* .0 4 .3 1* * .1 1 -.1 0 .0 9 .0 9 .0 3 (. 69 ) ve l 3. 56 1. 02 -.1 3 -.1 0 .0 5 .1 0 .3 8* * .3 0* * .2 2* * .3 2* * .2 7* * .2 8* * (. 43 ) es 2. 76 1. 41 .0 3 .0 1 .1 2 .0 5 .4 0* * .1 6 .0 5 .2 6* * .4 6* * .0 9 .2 7* * (. 67 ) tu re & nm en t 2. 71 1. 47 .1 8 .0 3 .1 5 .0 8 .4 1* * .2 6* * .0 6 .3 8* * .3 3* * .1 1 .3 6* * .4 3* * (. 59 ) er si on 4. 66 1. 36 -.0 3 -.1 2 -.0 2 .1 3 .1 0 .0 2 -.0 0 -.0 6 -.1 2 .0 6 .1 3 .0 1 .0 3 (. 74 ) bl en es s 4. 30 0. 96 .2 1* .2 8* * .0 0 .0 7 .1 2 .1 8* .0 3 .2 6* * .2 1* * -.0 1 .0 0 .0 6 .1 2 -.0 3 (. 36 ) en ti ou sn 5. 21 1. 15 .1 6 .3 1* * .0 5 .0 0 .1 5 .1 0 -.0 9 .1 2 .3 0* * .0 6 .0 4 .0 3 .0 1 -.0 8 .2 0* (. 18 ) ic is m 3. 14 1. 30 -.0 7 .1 3 - .21* * -.1 9* -.0 6 .0 2 -.1 3 .0 5 -.0 6 .0 2 .0 5 .0 9 .1 1 -.1 3 -.1 0 -.2 0* (. 64 ) ss 5. 27 1. 02 -.0 6 -.1 5 .0 7 .1 7* .2 5* * .2 0* -.0 4 .0 8 .0 2 .2 0* .3 3* * .2 9* * .2 6* * .2 6* * .0 5 -.0 3 -.0 6 (. 32 ) it y 3. 24 1. 50 .0 1 .1 3 .1 2 -.0 6 .2 7* * .2 0* .2 3* * .1 7* .1 2 -.0 2 .2 4* * .1 0 .2 8* * -.0 3 .1 1 .0 8 .1 0 -.1 8* (. 59 ) m en t 3. 67 1. 68 - .30* * .2 6* * -.0 3 -.1 6* .0 9 .2 6* * -.0 4 .0 9 .1 3 -.0 3 .2 1* .1 6 .1 3 -.0 2 .0 4 .0 8 .0 9 .1 3 .3 0* * (. 72 ) te nc e 3. 97 1. 58 -.0 8 - .24* * -.0 7 .0 3 .2 5* * .1 1 .2 1* * .1 2 -.0 0 .2 1* * .3 1* * .0 6 .1 2 .1 9* -.0 0 .0 0 .0 1 .0 6 .4 2* * .3 2* * (. 71 ) st ic at ion 3. 29 1. 69 -.0 3 .3 3* * -.1 3 -.0 4 .1 4 .1 8* -.1 7* .1 2 .2 5* * .0 7 .1 9* .1 1 .1 2 -.0 1 .1 8* .2 7* * .0 5 .0 8 .2 7* * .4 2* * .4 4* * (. 89 ) ne ss 4. 03 1. 34 -.0 6 -.1 3 -.0 3 -.0 0 .2 3* * .1 7* .2 0* .1 4 .0 3 .1 1 .2 7* * .2 0* .1 4 .1 4 .0 4 -.0 4 .0 4 .1 9* .3 6* * .2 9* * .4 3* * .1 8* (. 59 ) or re la ti on is s igni fi ca nt a t t he 0. 01 le ve l ( 2-ta ile d) re la ti on is s igni fi ca nt a t t he 0. 05 le ve l ( 2-ta ile d)

(37)

5.4 Regression analysis

The regression results of main effects and interaction effects of the control variables, consumer lifestyle, and consumer characteristics on fashion brand preferences are presented in Table 6. In model 1 the demographics age, gender, education, and income are used as control variables. This model shows the main effects of the control variables on the dependent variable brand preferences. Model 2 shows the main effects of consumer lifestyle and consumer characteristics on fashion brand preferences. It was predicted that both consumer lifestyle and consumer characteristics have a significant main effect on fashion brand preferences. Model 3 shows the interaction effects of consumer lifestyle and consumer characteristics on fashion brand preferences. It was predicted that the interaction between consumer lifestyle and consumer characteristics has a significant effect on fashion brand preferences.

Table 6 Regression results of main effects (model 1 & 2) and interaction effects (model 3)

Brand Preferences Model 1 Model 2 Model 3

B SE Beta B SE Beta B SE Beta Constant 3.248* .667 -.523 .959 2.462 4.896 Age -.143 .149 -.109 -.170 .132 -.130 -.174 .132 -.133 Gender .305 .213 .145 .173 .191 .082 .181 .192 .086 Education .006 .069 .008 -.031 .061 -.046 -.030 .062 -.044 Income .014 .094 .017 -.031 .084 -.038 -.033 .084 -.041 Consumer Lifestyle (CL) .685 .160 .406* -.097 1.268 -.057 Consumer Characteristics (CC) .363 .195 .177 -.284 1.059 -.139 Interaction CL*CC . .169 .271 .647 R² .027 .260 .263 Adjusted R² -.020 .209 .204 R² Change .027 .233 .003 Sig. F Change .718 .000 .535

Note: Dependent variable is brand preferences. N=152. *p <.01

The first model does not show any main effects between the control variables and the dependent variable fashion brand preferences. This corresponds with the correlation analysis

(38)

conducted before, where also no correlations were found between the control variables and the dependent variable fashion brand preferences.

The second model shows that the independent variable consumer lifestyle has a significant main effect on fashion brand preferences (β = .406, p < .01). This indicates that a consumer’s lifestyle does influence consumers’ preferences towards specific fashion brands, verifying the first hypothesis. This also corresponds with the correlation analysis, where it

was found that consumer lifestyle is a moderate predictor of fashion brand preferences (r = .35, p < .01). However, model 2 shows that the moderator variable consumer

characteristics has a positive effect on fashion brand preferences, but unexpectedly not in a significant way (β = .177, ns). This means that consumer characteristics do not necessarily influence consumers’ fashion brand preferences. This does not correspond with the correlation analysis, where it was found that consumer characteristics is weakly correlated with the dependent variable fashion brand preferences (r = .23, p < .01).

Also unexpectedly, the third model does not show a significant interaction effect between the variables consumer lifestyle and consumer characteristics on fashion brand preferences (B = .169, ns), not verifying hypothesis 2 the moderating effect of consumer characteristics on the relationship between consumer lifestyle and fashion brand preferences. This finding indicates that consumer characteristics do not necessarily moderate the relationship between consumer lifestyle and consumers’ preferences towards fashion brands as predicted before.

Furthermore, model 2 (R2 = .260) and model 3 (R2 = .263) explain more variance than model 1 does (R2 = .027). This indicates that model 2 explains 26% of the variance by the independent variables consumer lifestyle and consumer characteristics, and control variables and that model 3 explains 26,3% of the variance. However, the quality of model 2 is better than the quality of model 3, since the R Square Change (R2 Change = .233) leads to a

(39)

significant F Change (Sig F Change = .000) and the Adjusted R Square increases strongly to .209. Therefore, it can be concluded that the regression model improves when the independent variables consumer lifestyle and consumer characteristics are included, but not necessarily when the interaction variable was included.

Consumer lifestyle and brand preferences : investigating the impact of consumer lifestyle on preferences towards fashion brands