Who are they?

Who are they?

Communication of brand personality by business schools in

an online environment

Author(s): One-year Master Student Arber Tahiri, 881105 One-year Master Student Samira Youkhanna, 890328

Tutor: Dr. Rana Mostaghel

Examiner: Dr. Sarah Philipson

Subject: Brand; Brand Personality;

MBA; European Business Schools: Content Analysis Level and semester: 4FE02E,

Acknowledgements

The authors would like to take this opportunity to give a special thanks to the individuals who helped and contributed with their valuable insights and ideas in the writing of this study. With humble gratitude, the authors of this study would like to extend their thanks to Dr. Rana Mostaghel, whom was kind enough to inspire us from the start of the thesis with her passion for marketing and considerate guidance. The thesis has been an instructive and inspiring process from idea to the completion.

The authors would also like to give a special thanks to Dr. Sarah Philipson, whom not only was kind enough to manage to spare time to assist us during the stressful period of writing but also contributed with her professional knowledge about researching which has given us the chance to better understand the research world and the art of writing a thesis. In addition, we want to send our kindest regards to our opponents and Magnus Hultman for being a source of information. Their comments and commitment to us are highly appreciated.

Växjö, 20120528

____________________ ____________________

Samira Youkhanna Arber Tahiri

Abstract

Title: Who are they: a study about Communication of brand personality by business schools in

an online environment

Course: 4FE02E, Master Programme Degree Thesis, 15 ECTS Authors: Arber Tahiri

Samira Youkhanna

Tutor: Dr. Rana Mostaghel Examiner: Dr. Sarah Philipson

Introduction: MBA programs have increased from having 3200 awarded degrees during the

50‟s to 1695 business schools offering different types of MBA. Researchers claim that the growth of MBA programs may have changed human perception of the offered program, that MBAs have gone from being a first class ticket to a successful career to a more accommodated service. Brand and brand personality is another discussed aspect. Researchers suggest that organizations should implement more strategic communication based on brand personality, on their websites to attract a large audience and clearly communicate what and who they are as an organization.

Purpose: The purpose of this study is to examine the brand personality communication of

European business schools MBA Programs in an online environment.

Methodology: The study includes a comparative study were a correspondence analysis was

conducted. Aakers (1997) model of brand personality dimension was used as an inspirational tool and randomly selected two samples; Cluster one, consisting of five Top European business schools according to Financial Times, and Cluster two, five general European schools.

Conclusion: According to findings, there are some areas such as the usage of Aaker‟s brand

personality dimensions and their synonyms that are different between top ranked European business schools and other business schools in Europe. Also, top European business schools appear not to be closely associated to Aaker‟s brand personality traits whilst other European schools are closely knitted and communicated clear brand personality dimensions.

Table of content

1. INTRODUCTION ... 9 1.1 BACKGROUND ... 9 1.2 PROBLEM DISCUSSION ... 11 1.3 PURPOSE ... 13 1.4 DELIMITATIONS ... 13 2 THEORETICAL FRAMEWORK ... 14

2.1 ONLINE MARKETING AND COMMUNICATION ... 14

2.2 BRAND ... 15

2.3 BRAND PERSONALITY ... 16

2.4 SOCIETAL AND SELF-CONSTRUCTION REALITY OF BRAND PERSONALITY ... 18

2.4.1 BRAND PERSONALITY FOR CUSTOMERS ... 19

2.5 THE “FIVE DIMENSIONS” ... 20

2.6 CRITICISM AGAINST AAKERS BRAND PERSONALITY DIMENSION ... 21

2.7 STATE OF THE ART ... 23

2.8 SCIENTIFIC QUESTION ... 25

3. METHODOLOGY ... 26

3.1 SCIENTIFIC APPROACH ... 26

3.2 RESEARCH DESIGN... 27

3.2.1 QUANTITATIVE AND QUALITATIVE RESEARCH ... 27

3.3 POPULATION AND SAMPLING ... 27

3.3.1 POPULATION ... 27

3.3.2 SAMPLING ... 28

3.4 RESEARCH DATA COLLECTION ... 29

3.4.1 DATA COLLECTION IN PRECISE STEPS ... 30

3.5 DATA ANALYSIS METHOD ... 30

3.5.1 CONTENT ANALYSIS ... 30

3.5.2 CORRESPONDENCE ANALYSIS ... 31

3.6 OPERATIONALIZATION ... 32

3.6.1 STEP ONE AND TWO: AAKER’S (1997) FIVE BRAND PERSONALITY DIMENSION AND BRITANNICA ENCYCLOPEDIA ... 33

3.6.2 STEP THREE: DESIGNING THE DICTIONARY ... 34

3.6.3 STEP FOUR: WORKING WITH WORDSTAT ... 34

3.7 VALIDITY & RELIABILITY ... 35

3.7.1 VALIDITY ... 35

3.7.2 RELIABILITY ... 37

4.1 DISTRIBUTION OF FREQUENCY AND BRAND DIMENSIONS ... 39

4.1.1 DISTRIBUTION OF FREQUENCY OF BRAND PERSONALITY DIMENSIONS FOR CLUSTER ONE AND TWO ... 42

4.2 TWO DIMENSIONAL CORRESPONDENCE MAP ... 44

4.3 COMMUNICATING AND DELIVERING INFORMATION ... 46

5. ANALYSIS ... 48

5.1 DISTRIBUTION OF KEYWORDS AND FREQUENCY ANALYSIS ... 48

5.2 TWO DIMENSIONAL CORRESPONDENCE MAP ANALYSIS ... 49

5.3 COMMUNICATING AND DELIVERING INFORMATION ... 52

5.4 SUMMARY OF ASSOCIATIONS BETWEEN THE CASES AND THE BRAND PERSONALITY DIMENSIONS ... 55

6. CONCLUSIONS ... 57

6.1 THEORETICAL AND MANAGERIAL IMPLICATIONS... 58

6.2 LIMITATIONS ... 59 6.3 FURTHER RECOMMENDATION ... 59 7. REFERENCE LIST ... 61 7.1 SCIENTIFIC ARTICLES ... 61 7.2 ELECTRONIC SOURCES ... 73 8. APPENDIX ... 75 APPENDIX 1 ... 75 MBA Programs ... 75

Europe and MBA programs ... 77

Fulltime vs. Part-time MBA programs ... 78

One vs. two year MBA ... 78

Business schools and their MBA programs as brands ... 79

APPENDIX 2 ... 81

APPENDIX 3 ... 82

APPENDIX 4 ... 84

APPENDIX 5 ... 91

APPENDIX 6 ... 98

London Business School ... 98

IMD Business School ... 102

IE business School ... 104

University of Oxford: Said ... 106

HEC Paris ... 111

Kingston Business School ... 115

Trinity College Dublin ... 119

University of Geneva ... 123

Paris School of Business ... 127

Sabanci University... 131

APPENDIX 7 ... 133

APPENDIX 8 ... 135

Outline

C

HAPTER ONE: INTRODUCTION

The chapter introduces the thesis and its background, giving the reader a pre-understanding about the subject and problem area and finalizing with a purpose for the study.

C

HAPTER TWO: THEORETICAL FRAMEWORK

This chapter gives an exploratory theory background for the thesis. It includes definitions and further in depth information about subjects such as „branding‟, Aaker‟s brand personality dimensions‟ and the different aspects constructed reality.

C

HAPTER THREE: METHODOLOGY

The methodology chapter presents the progression of the deductive study and how information for the empirical framework has been collected and used. Content analysis have been the preferred method to gather necessary information about the business schools and their MBA programs by randomly selected European business schools from two Clusters; one presenting top European business schools according to FT and another one presenting most European business schools in Europe.

C

HAPTER FOUR: EMPERICAL DATA

This chapter presents gathered empirical information form the software program WordStat. It describes the total frequency of distribution of keywords and how the personality dimensions are connected to each European Business School. The empirical study also provides a chi-square test to prove if there is any significant relation between the schools and brand personality traits according to Aaker (1997).

C

HAPTER FIVE: ANALYSIS

C

HAPTER SIX: CONCLUSION

The chapter concludes the results, which are highlighted in the analysis and answer the research questions in the study.

C

HAPTER SEVEN: FUTURE RECCOMENDATIONS

1. Introduction

The thesis starts by providing the reader with a background including the origin and development of MBA programs around the world. The introduction will then funnel down to a problem discussion, which will include in-depth discussions about different gaps in research of brand identity and business schools in an online environment. The discussion will eventually end with a purpose.

1.1 Background

Masters of Business Administration (MBA) programs (See Appendix I) give profits for most of the parent institutions, and provides well-trained business executives for organizations, and these are perceived to be the corporate elite in the area of management education (Pfeffer and Fong, 2002). Pfeffer and Fong (2002) claim there is no doubt that business education and MBAs enable great profits and are considered big business. Business schools spend a large amount of money on the advertisement and promotion of their MBA programs to strengthen the brand and attract a new target group of students.

In general, business education has grown worldwide. Only in Britain the number has risen from 20 in the early 1980s, to over 150 in 2005. In the US, MBA programs were nearly non-existent between 1955-1956, only 3200 MBA degrees was awarded. However, 1998-1999 nearly 102 000 MBAs was credited and by the fall of 2000 there was 341 accredited programs in the US. Business education has spread around the world, in places such as Asia and continental Europe (Hultman et al., 2007; Pfeffer & Fong, 2002). Businessschools.com claims to have 1,695 MBA Business Schools and Colleges in its database (Browse MBA programs, 2012). Hultman et al. (2007) claim in their study that there is a growth of 1500% in Internet trafficking in websites in 2005, and it is still generating 30,000 plus referrals per month since the time of year 2006.

high importance as numerous students use them. Also, the ratings provide a snapshot of the graded schools‟ reputation and are suggested not be ignored, but used to gain advantage over competitors (Opoku et al, 2007; Bickerstaffe, 2005; Friga et al. 2003).

Keller (2008) suggests the very important changes, namely the increased amount and the diversity of the communication options available to marketers. Reaching consumers has become a challenge, as there are emerging non-traditional media, promotion and other communication alternatives (Keller, 2008). Furthermore, one of many important challenges in marketing MBA programs is the importance of brand personality and how they are constantly communicated over the Internet the same way as in real life (Hogg et al., 2000; Aaker, 1997). Brand personality defines a set of individual personality attributes that can be applicable and relevant for brands, (Azoulay and Kapferer, 2003; Brown, 1991). Keller (2008) mentions that academics usually conceptualize brand identity and brand image as multi-dimensional constructs of which brand personality is an important component. Rowley (2002) concluded that online branding communication conveys a sufficient range of challenges and opportunities, one of them concerning exploration of some of the potential impacts of online channels (such as websites) on branding strategy. According to Pitt et al. (2007), visitors to websites tend to be highly involved and active in acquiring information from organizational websites, as the number of visitors that turn to organizational websites to gather information about a program tend to increase. Hence, it is therefore important for businesses schools to clearly and distinctly communicate brand personality on their own website to make them stand apart from competitors, Pitt et al. (2007).

1.2 Problem Discussion

Onlinemba.com (2012) claim that MBA degrees are considered as the world‟s most popular postgraduate degree and are considered as a first class ticket to a successful management career. On a business point of view, there is an increasing interest by marketing schools to understand and measure the symbolic meaning that consumers attribute to the represented brand personality of the MBA degree (Pitt et al., 2007). Hence, researchers conclude that there is a challenge in how consistent the communication of MBAs and business schools is in the online environment, and if it is done in the same manner as in the real world (Hogg et al., 2000; Aaker, 1997). Accordingly, it is not only problematic to choose the „right‟ personality elements for the brand, but also difficult to enable a consumer to develop a visible and unique representation of him or herself that matches with it.

From a scholarly perspective, one of the biggest challenges is the communication of the brand personality of an organization (Van Gelder, 2005; Ligas, 2000). Yet another challenge, studied by Perry and Bodkin (2002; 2000), is if the company does not have adequate resources, which leads to a limited engagement on the website. Further, many organizations are not convinced that they need to establish and maintain a sustainable website. In some cases they just do not know how to utilize it (Van Gelder, 2005). On the contrary, business schools, which have shown an interest in brand management, have also shown to have different conceptualizations of brands, including symbolic components (Keller, 1993, 2008; Van Gelder, 2005; Ligas, 2000). However, according to Austin et al. (2003), a few valid measurements have been successfully developed, the online environment being one of the biggest challenges, according to studies by Keller (1993; 2001; 2008). Many organizations do not implement the full range of marketing communication on their websites, because they do not have the financial (money) and human resources (employees) to build and maintain effective corporate websites (Hultman et al., 2007; Van Gelder, 2005).

However, academic researchers are deserting brand personality as relevant and vital for business schools in their communication (Mullin, 2000). Further, academic literature are largely neglected the growing importance of online communications, as part of corporate identity management. This has had a negative impact on the communication of the brand personality in the online environment (Hultman et al., 2007), as it limits rather than fostering marketing communication. In agreement with Pitt et al. (2007), Keller (2008), claims that academics ignore the fact that a business school‟s website is a very important element of its corporate identity, since it explains what the school is about and „who they are‟. Pitt et al. (2007) suggests in their study that brand personality has often been put in relation and reference to products, corporate brands or countries, and not to brand communication online, which has shown to have a negative impact on communicating business schools‟ brand personality, as they will face future difficulties in differentiation and creation of strategic advantage (Hultman et al., 2007; Van Gelder, 2005).

1.3 Purpose

The purpose of this study is to examine the brand personality communication of European business schools MBA Programs in an online environment.

1.4 Delimitations

2 Theoretical framework

This chapter identifies theories such as online marketing communication and branding. Further, previous research of concepts such as Aaker’s (1997) brand personality dimensions will be presented, as well as information about MBA programs in Europe. The line of argument is; what are the different types of brand personalities and how are they conveyed in brands of business schools’ that offer MBA programs? The theories will eventually funnel down to a research question.

2.1 Online marketing and communication

Keller (2001) defines the term marketing communication as the way a firm informs, persuades, insight and remind consumers (directly or indirectly) about its brands and products. Marketing communications has faced changes and challenges over the years, including design, evaluation and implementation of marketing communication programs (Keller, 2001; Gallagher et al., 2001). Another change is the increase of the number of diversity of communication options that are available for marketers in a way that was impossible in earlier days (Keller, 2001). Analogue communication is decreasing in a fast pace as digital media is taking over alongside with the development of the World Wide Web (Bickerstraffe, 2005; Rowerly, 2004). According to Gallagher et al. (2001), the new changes give the audience much more control over exposure to material on the web and they can chose which website they want to enter and explore.

Websites has become a tool to use when representing status symbols of organizations (White and Raman, 1999). Also, websites tend to be brand carriers and explain the extension of the organization‟s operations (Palmer and Griffith, 1998). Most organizations use a variety of Internet marketing communication tools (such as websites) to communicate with target customers, and to build customer relationships (Perry and Bodkin, 2000, 2002; Gallagher et al., 2001).

customer service, and streamline online reservations to please target costumers and recruit new ones (Baloglu & Pekcan, 2006; Gallagher et al., 2001). According to Austin et al. (2003) organizations emerge their brands by presenting it, as well as presenting the organization to the public e.g. via television advertisements, media coverage and press releases, the Internet, personal familiarity with organization location and/or information obtained from friends, family and acquaintances.

2.2 Brand

Kapferer (2008) suggest that brand identity is a brand that is specified and presented to an audience by a firm for its target group. According to Kapferer (2008), the definition of brand is hard to define as authors define the term in their own way. He also state that defining a brand gets more complicated when it comes to measuring the strength of brands, however, he believes that the focus should instead be put on what is commonly called brand equity. According to Kapfrerer (2008) there are different views of branding and brand equity. One is customer based and focuses on the relationship between the customer and the brand (e.g. loyalty, attachment and willingness to buy and rebuy based on customer beliefs of superiority and evoked emotions). The second aims towards the productions of measures in terms of money (Kapfrerer, 2008).

brand in their own way without the fear of being copied by others, De Charnatony and Riley (1998).

Stakeholders as suppliers, government and shareholders and have their own view of the company's brand. However, companies have two most important stakeholders, the clients and the employees. A simplified representation of the process is found in the model below. The model consists of an explanation where the brand is the interface between business and how the consumer interprets it De Charnatony and Riley (1998).

Figure 2.2.1: Defining a Brand, De Chernatony, L. and Riley, D. (1998) pp. 428

De Charnatony and Riley (1998) claim that companies may obtain information by monitoring customers‟ perception of a brand to reduce the gap between the company and the customer's attitude towards the brand. The company also has the opportunity to reinforce the brand and make it more identifiable and more relevant for the consumer.

2.3 Brand personality

Ryckman (1997) divides personality in five categories:

i. The Psychoanalytic perspective which is about the biological in nature of personality is also served as an underlying bases for certain kind of personality

ii. Trait perspective is about dispositional factors that regularly and persistently determine

conduct in a variety of everyday situations;

iii. Cognitive perspective assumes that people‟s personality is never completely

determined; people are always changing and always free [a bounded freedom] to reinterpret their experiences in idiosyncratic ways;

iv. Existential perspective postulates the existence of an innate growth that moves individuals toward realization of their potentialities if environmental conditions are ideal and

v. The social behaviourist perspective assumes that most of our behaviour is learned and purposive; we are guided by our motives to attain certain goals.

Aaker (1997) state that brands, like individuals, are able to develop personalities that are similar in matter of characteristics. Plummer (1984, 2000), suggest that a strong brand personality is irreplaceable and valuable to build a brand and brand equity, and cannot be easily copied.

Aaker (1997) define brand personality as "the set of human characteristics associated with a brand." (Aaker, 1997:347). Further, an individual can be seen as brand, for instance famous artists are used in promotional aspects by representing companies and organizations, according to Solomon et al. (2010). In addition, Keller (1993) state that brand personality is something that serves as a symbolic or self-expressive function. Ghodeswar (2008) conclude that companies may use brand personalities to make it easier for them to reach a preferable audience. As a result, companies faces no difficulties differentiating themselves from competitors. Also, companies using brand personality enables a better image and contributes to enhanced products by implementing brand personality in brand communication (Ghodeswar 2008).

customers to see a connection between brand and their human characteristics in order to interact and understand the brand they are involved with. However, McCrachen (1989) claim that companies can affect brand personality by using famous people in different ad campaigns. He suggests that some companies‟ may use celebrities to represent a product or service to give the brand a direct personality (McCrachen, 1989). It enables a chance for the audience to identify and relate to the celebrity that is connected to the corporate brand communication.

2.4 Societal and self-construction reality

of brand personality

Bhattacharya and Sen (2003) mention in their study that brand identity can be both socially and self-constructed. Social identity ”theory posits that in articulating their sense of self, people typically go beyond their personal identity to develop a social identity. They do so by identifying with or categorizing themselves in a con-textual manner as members of various social categories (e.g., gender, ethnicity, occupation, sports teams as well as other, more short-lived and transient groups)” (Bhattacharya & Sen, 2003:77). Further, Arinaitwe (2006) suggest that actors such as firms, suppliers, supplementary organizations, the public, and customers can make the social construction. In addition, Bhattacharya and Sen (2003) mention that the bases of social constructed brand identification are for consumers to identify, or feel a sense of belonging to an organization.

2.4.1 Brand

personality

for

customers

Brand can be associated with the personality traits, means that it can provide self-expressive or symbolic benefits for the customer (Aaker 1999). Consumers identify themselves with a certain brand that reflects and reinforce their self-identity. If the product or service fit with the image a decision would be made (Solomon et al 2010). E.g. the customer buys a Harley- Davidson motorcycle, because it has an image of masculinity and strength, Aaker (1997).

If a company is interested in building a preference for a product they can do it in six different ways:

i. Need association; a particular need of the product, which the customer can associate with;

ii. Mood association; letting the customer know the process of the product production.

What link it has with different aspects, example special moods;

iii. Subconscious motivation; manipulating the customers mind through suggestive symbols, without the customers knowledge;

iv. Behavior modification; manipulating the customer through different rewards and signals;

v. Cognitive processing; is about creating cognitive and perceptual barriers through positive attitudes and

vi. Model emulation that imitates the introduced idealized social lifestyle models. (Alreck & Settle, 1999).

According to Aaker (1997), there are two ways of finding the relationship between brand and human personality, which can have an impact on consumers‟ preferences. The two ways are:

i. Ad hoc scales: are normally composed of a set of traits ranging from 20 to 300.

The scales are not theoretical in nature; instead they have often been developed for the purpose of specific research. As a result, key traits may be missing from such scales. The traits that are selected are often chosen arbitrarily, which casts doubt upon the scales‟ reliability and validity (Aaker 1997).

ii. Aaker (1997) state that the ones that are more theoretical in nature are the second types of brand personality scales. However, they are based on human personality scales that have not been validated in the context of brands. Therefore, some factors of human personalities can be reflected, but not all. The validity of brand personality scales has become a matter of question, Aaker (1997) state, “if unequivocal results are to emerge, consumer behavior researchers must develop their own definitions and design their own instruments to measure the personality variables that go into the purchase decision” (Aaker, 1997:348).

Aaker (1997) discusses how distinct dimensions can isolate brand personality by considering brand personality as a uni-dimensional (value being measured wholly in terms of its truth) construct, and by showing how diverse types of brand personalities can be distinguished. As a result, there are numerous ways in which a brand personality construct can affect individual‟s preference and how this can be better understood.

2.5 The “Five dimensions”

Aaker´s scale of brand personality measures human brand personality and the relation between customers and corporate brand, Aaker (1997).

Further, Caprara et al. (2001) state, “The tradition in personality psychology maintains that over time, languages develop a list of adjectives that describe the most important personality differences between people. Factors analyses among thousands of personality attributes have revealed that, in many different languages, most adjectives describing individuals can be grouped into one of the so- called big five” (Caprara et al., 2001:378).

Further, Goldberg (1993) claim that the five dimensions, commonly known as the “Big Five”

found to be easily evoked by, limited facts of adjectives that do illustrate a high loading in the desired aspect and low loadings in the remaining factors.

Figure 2.5.1: A Brand Personality Framework

Figure 2.5.1: A Brand Personality Framework (Aaker, 1997, pp.352)

According to McCrae & Costa, (1997) Big Five researchers deliberately exclude gender and social class, however, Aaker (1997) provide another perspective of brand personality, sincerity stated as the first of five factors. This is a dimension that encapsulates concepts like acceptance and warmth. The second one is excitement, which can be connected with energy, sociability and activity. Competence is the third factor and this is associated with security, responsibility and dependability. Further we have sophisticated, this gets associated with concepts like sexiness, glamour and aspiration. The last one is ruggedness and it‟s associated with masculinity, strength and glamorized ideals. Figure 2.5.1 show different concepts that are recognized with each of the five factors (Aaker, 1997).

2.6 Criticism

against Aakers brand

personality dimension

i. Loose brand personality traits: The first criticism aims for the loose definition of brand personality. Further, it is something that embraces different characteristics beside personality, for instance sex, age etc. (Azoulay & Kapferer, 2003; Bosnjak, Bochmann, & Hufschmidt, 2007). According to Geuens et al (2007) this is something that induces a construct validity problem and leaves investigators and practitioners uncertain of what they have truly measure; the perceived brand personality or perceived user characteristics?

ii. Brand personality as a non-generalizability factor: The second criticism involves the

non-generalizability of the personality trait factor structures for analyses at the respondent level e.g. for a specific brand or product category (Austin, Siguaw, & Mattila, 2003). Cited by Geuens et al. (2007) “Aaker (1997) conducted all analyses on data aggregated across respondents (for between-brand comparisons), she actually removed all within-brand variance, which led to factor analysis results that are exclusively based on between-brand variance” (Geuens et al., 2007:97). As a backdrop, the framework does not cover generalizations in which analyses are required at the individual brand level and/or situations where consumers are an element of differentiation, Geuens et al. (2007).

iii. Brand personality as a non-replicable tool: The third criticism explains how the five

factors (personality traits) within brand personality are non-replicable across cultures (Azoulay & Kapferer, 2003). For example, Aaker et al (2001) gave some examples of different country‟s that was studied and it was known that in a study that she made in Spain, only three of five factors (personality traits) were used in terms of studies. Further, in Japan she made same study and in that case four of the five factors were used (Geuens et al (2007).

Further, the personality characteristics scale has served as a brand personality measure in many studies, however characteristics are proved to be robust in several studies (Aaker, 1997, 1999; Aaker, Benet-Martinez, & Garolera, 2001; Kim, Han, & Park, 2001).

(intellectual, imaginative, independent-minded). Further, a number of researchers found other different sets of hundreds of personality characteristics that are related to the recurrent five personality factors (Costa & McCrae, 1992; Goldberg, 1992; Norman, 1967). Hence, the characteristic factors are not always identically labeled. Neuroticism/Emotional stability has emerged as Emotionality and Affect whereas Openness/Intellect has appeared as Imagination, Culture, Rebelliousness, and Unconventionality. Further, researchers have suggested Conscientiousness as Responsibility (John & Srivastava, 1999). Conferring to what is stated above, the general outline of the Big Five seem to appear in most studies, more often in the cross-national ones, and the evidence is at least convincing for the Openness factor, John & Srivastava (1999).

2.7 State of the art

This study presents different theories to give the reader various views on the chosen subject. Theories from Aaker (1997), Ryckman (1997), John and Srivastava (1999), Keller (2001), Bhattacharya and Sen (2003), and Geuens et al (2007) have been the base for this thesis.

hundreds of times by researchers whom has been cited many times and made well accepted empirical studied strongly validated by other researchers. Therefore the sources have a strong validation.

The concept of brand personality was based on Aaker (1997) and Ryckman (1997). Geuens et al (2007) claim that Aakers (1997) brand personality dimensions have loose ends in many aspects, such as personality traits, generalization, and replicability and it is therefore a non-validated source (Azoulay & Kapferer, 2003; Geuens et al., 2007; Bosnjak, Bochmann, & Hufschmidt, 2007). Geuens et al. (2007) has limited citations, however their study is validated due to the amount of empirical studies connected to the cited study and can therefore be seen as an accepted theory with a limited but validated acceptance. Aaker (1997) is an accepted theory in the area of brand personality, but not dominating. Hence, her article has many quotes from researchers, such as Fournier (1998), Pitt et al. (2007), Opoku et al. (2007) and Van Gelder (2005), have all been well-cited and well accepted due to its validated reference, based on empirical studies made by other researchers whom has used the cited authors in their empirical investigation. Ryckman (1997) has presented a book and not an article however it is accepted in the research world by other authors who has cited the source. The source as such is not validated because the majority of the cited sources are books and based on other sources.

Keller (1993) and Bhattacharya and Sen (2003) add new perspectives to brand personality; so-called self and social constructive reality (Keller, 1993; Burr, 1995; Bhattacharya and Sen, 2003). Burr (1995) have gathered different sources and cited other authors in his book however it is accepted in the research world by other authors who have cited the source great amount of times. The source as such is not validated because the majority of the cited sources are books based on other sources however it will be used for information and clarification purposes in the theory chapter. Further, the study will approach a self-constructed reality, according to Keller (1993), which is a dominated source, cited by many researchers who have validated it by presenting various of empirical research in other areas as well as branding.

another aspect to the big five dimensions (extroversion, agreeableness, conscientiousness, emotional stability, and openness). However, Aaker‟s (1997) presented personality dimensions (sincerity, competence, sophistication, excitement and ruggedness) will instead be used as a basis for the study, as her work has inspired the majority of the research on brand personality to date as well as it is well-cited and well-accepted in the research world. Researchers has performed empirical investigations and studies in areas such as branding, CMR, customer loyalty aimed to both conceptualize and validate the brand personality theory of Aaker (1997). To construct a brand personality scale, Aaker (1997) started from Big Five items, but completed them with, amongst other things, socio-demographic characteristics. Consequently, whereas Big Five researchers deliberately exclude gender and social class (McCrae & Costa, 1997), Aaker includes feminine, upper class, young, etc. Also, the structured personality traits are validated in various studies/researchers that are well-cited and accepted in the research world (Aaker, 1997, 1999; Aaker, Benet-Martinez, & Garolera, 2001; Kim, Han, & Park, 2001). With these arguments Aaker (1997) is an accepted source and is validated by other researchers. Aaker (1997) has been used as a base for theory, and the authors suggest her sources to be the base for the empirical framework in the study.

2.8 Scientific question

Websites may be similar to each other, both in appearance and function. Therefore, brand personality can be a strategic tool to use when wanting to differentiate from other competing sites, Kim et al., 2001. Perry and Bodkin (2002) claim that it is necessary for organizations to use clear and integrated marketing communications website strategy to differentiate due to the incensement in using websites to foster relationships with customers. One can believe that business schools may use the advantages of websites to differentiate themselves from competitors when communicating their brand personalities. This is due to websites and the ways that differ from one organization to another (Berthon et al., 1998; Kiani, 1998). The authors would like to investigate what the exact reasons are behind the similarities/dissimilarities, in what way the two Clusters differ from each other and „what the similarities/dissimilarities mean‟. Therefore, the following question will be:

3. Methodology

This chapter will discuss the methodology of this study. First, scientific approach then the type of research strategy, research design as well as research method will be described. Second, the population and sampling used for the data collection is presented. We will end this chapter by presenting the operationalization, testing quality criteria and content analysis.

3.1 Scientific approach

To find theories that are relevant to this paper the authors used a deductive method, which refers to drawing general conclusions that are founded on the basis of the theory (Bryman and Bell 2005). Bryman and Bell (2005) state that the deductive approach is based on theory, where the theory represents the most ordinary sense of how the relationship between theory and practice is, in the societies point of view and do not describe if something is wrong or right (Bryman & Bell, 2005). Deductive approach differs from an inductive, which is based on the empirical data and were conclusions are drawn from, instead of the theoretical relationship and foundation (Bryman and Bell, 2005; Holmes & Solvang, 1997)

3.2 Research design

3.2.1 Quantitative

and

Qualitative

research

Qualitative research is used to gather an in-depth understanding of human behavior and the reasons that leads to such behavior. Qualitative research process involves questions that are more open than the quantitative method, and investigates the how and why questions of decision-making. Further, it´s more focused on the respond from the respondents and smaller but more focused samples are more often needed than large samples. Qualitative research focuses on meaning, process and understanding resulting in a descriptive study (Bryman and Bell, 2010).

This study was conducted by using a qualitative data collection. The data was analyzed by words from different, existing MBA business schools websites in Europe, which means that secondary data was used. The qualitative data was rich in the content and provided us with a detailed understanding. The authors feel that this was a vital first step, as investigators and marketers try to form this promising and interesting zone. The theory that was used led the research to a research problem and this problem got the best response by different websites.

Later on the questions got processed in charts and tables and became the underlying basis of the empirical data. This study is based on quantitative analysis of qualitative data presented by European business schools.

3.3 Population and Sampling

The main purpose of our sampling was to gather enough cases help us to study the phenomenon of brand personality communication in our target population.

3.3.1 Population

programs, such as part-time, full-time and online MBAs to generalize rather than limiting the study. As stated in the background, there has been a large increase in the amount of MBA programs in the world, hence this study did not have the opportunity to investigate all schools that that offer MBA programs, instead the focus was on European business schools as state din the delimitations chapter. The authors wanted to find out if there were any differences or similarities between schools MBA online communication.

3.3.2 Sampling

There are two types of sampling, random and non- random selection, a random sample was selected for this study, which means that the gathered business schools providing MBAs had the same chance to participate in the study (Bryman and Bell, 2010). Therefore, the possibility of generalization in this study became quite high (Bryman and Bell, 2010). Using convenience sampling will often generate missing data in relation to the population, Christensen et al. (2008), which indicated the choice of European business schools that were included in the Financial Times top 100 ranking system in the world, 2012. We also decided to use MBA programs that are in the list of topmba.com. The authors started by listing all 100 MBA programs from FT list and then piled them in to a separate table, see Appendix 3. The second stage included the collection of all the MBA programs in Europe, provided by topmba.com, listing them to provide a second table, see Appendix 4.

To randomly select five MBA programs from FT list, we used “Monte Carlo Simulation”. Cluster one was based on the five MBA programs from the Financial Times list whereas Cluster two originates from www.topmba.com list. The five schools that were selected in Cluster one were: London Business School, UK; IMD, Switzerland; IE Business School, Spain; University of Oxford: Said, UK; HEC Paris France. Cluster two included: Kingston Business School, UK; Trinity College Dublin, Ireland; University of Geneva , Switzerland; Paris School of Business, France; Sabanci University, Turkey. For interpretation purposes, two tables were made.

Table 3.3.2.1: The Initial and Reconsidered Rankings for Cluster one in this Study

Initial Ranking Our Ranking Name

4 1 London Business School

8 2 IE Business School

13 3 IMD

18 4 HEC Paris

Table 3.3.2.2: The Initial and Reconsidered Rankings for Cluster two in this Study

Initial ranking Our Ranking Name

24 1 Kingston Business School

123 2 Paris School of Business

178 3 Trinity College

230 4 Sabanci University

238 5 Executive program (HEC)

Cluster one was implemented in the sampling program for random selection of five MBA schools from a total of 26 schools. The Cluster presents five top European schools according to Financial Times. The presented schools have been ranked by the Financial Times since 2009 and are considered to be among the major players of the “elite” MBA Business Schools. Same method was used to select the schools in Cluster two, which had 243 in total. Cluster two schools have been listed by the www.topmba.com and are considered to be among the average MBA Business Schools in Europe. Note that the authors have given each scholar a new ranking number to make the empirical data collection and presentation easier to follow for the reader. Also, the two Clusters have been used to make it easier for the authors to separate the top ranked schools from the general, which was very useful during the time of writing.

The authors of this study believed that this was a decent method of basing the study on. It was believed that using the MBA programs that are included in the Financial Time and topmba.com list is a noble and necessary way for the study to present comparisons between elite and general schools.

3.4 Research data collection

3.4.1 Data collection in precise steps

The authors started in the gathering process 2012-05-03 by finding all the relevant webpages. The data of this study was based on official websites form each of the 10 MBA programs. To find the right website address, the Financial Time list of top MBA programs and the list of topmba.com was used. These two scores provided us with the right URLs. We primarily took all the information from the main pages and then copied all information that was available and needed for the study. Hwang et al (2003), state that homepages has good information but all information is not in the form of words and could also be communicated through different pictures, symbols etc. and may not be processed in the study. Consequently, we carefully removed information and links that was connected with ads, symbols, graphics, videos etc. To increase the reliability of the research, we tried to collect similar data from all the ten different MBA programs websites. However, it was not possible to find identical data for each website since every MBA program had different constructions on their webpages. Another possible choice to conduct the gathering of information could have been to include the same amount of information from each school. However, this would have created a risk for the loss of important information. Due to the fact that websites may change during different time aspects, the authors collected all data during four days, with the start on a Thursday and ending on a Sunday, because we wanted to accomplish a point- in- time assessment of existing MBA websites.

3.5 Data analysis method

3.5.1 Content analysis

study did also choose to rely on words in terms of content analysis with help of computer supported-content analysis WordStat software program, which according to Stone (1997) and Péladeau (2003) has been „on the word‟ as unit of analysis rather than on sentences (Roberts, 1997a).

Word unit analysis is of interest for the authors, not to say that other methods are irrelevant, motivated by the massive amount of words and text to manage. The authors believed that 1) the language organizations use mirrors what and/or who they are, concerned with, attending to or focus on; 2) that the frequency of certain usage of words on a website indicates of how the sender of a message focuses on a particular brand personality dimension. The authors also assumed that the used words in the chosen websites would reflect what the business schools consider important attributes or characteristics of the organizations and program.

The correspondence analysis was based on the software program called WordStat, which was the relevant program used in the study. According to provalisresearch.com WordStat can perform:

i. Analysis on text arenas in several formats, and also in long documents; ii. Process texts reducing words to canonical form;

iii. Univariate frequency analysis and presents results in matrix form;

iv. Bivariate comparison between any textual field and any nominal and ordinal variables, for instance, age or sex of the respondents.

The four performance areas above were also included and performed in the study and the outcome (tables and figures) were included and used as basis for the empirical framework (see Chapter 4). As a result, the gathered textual information from the ten selected MBA programs websites was transformed into an analyzable format and imported into WordStat. However the authors performed a content analysis of the ten selected business school websites in Europe, by analyzing the collected data from WordStat to identify relations between the schools and the brand personality dimensions.

3.5.2 Correspondence analysis

1994; Greenacre, 2000). However, correspondence maps may be used to evaluate the positioning and distinction between websites and brands, Berthon et al., (1997b; 2001a). This study has aimed to include a correspondence analysis in terms of plots/maps in order to help position the various websites and brand personalities in relation to another. According to Shanka et al., (2005), correspondence analysis is a mapping technique and thanks to WordStat one can visually display unique characteristics of groups and sub-groups in a high resolution line and bar charts and through two- Dimensional correspondence analysis maps (which can also be spotted in chapter four and five). Bendixen (1995), explains a correspondent as analysis cross-tabulates data method that is used for conversion to a joint space map by using the chi-square value for each cell. Accordingly, this kind of method is not new in the area of perceptual map construction, Greenacre, (1989); Lebart et al., (1984); Hoffman and Franke, (1986); Hair et al., (1998), in the area of Clustering of brands (Bendixen, 1995; Hair et al., 1998) and choice evaluations (Shanka et al., 2005).

Using a correspondence analysis was a sharp choice for this thesis as it is exploratory in its nature and the analysis method would be useful since it uncovers structural relationships between different variables (Inman et al., 2004). Also, the correspondence analysis enables simple interpretation of data that otherwise would be difficult to comprehend thanks to its graphic nature (O‟Brien, 1993). Therefore, the authors believe that a correspondence analysis would be appropriate to use for exploratory data analysis (Hair et al., 1998). The outcome of the correspondence analysis includes graphics presented in Appendix 7 and Appendix 8 as well as in the empirical chapter to make it easier for the reader to follow.

3.6 Operationalization

3.6.1 Step one and two: Aaker’s (1997)

five brand personality dimension

and Britannica Encyclopedia

First, the authors collected and complied synonyms with Aaker‟s (1997) five personality traits. Aaker (1997:354) introduced a table with a set of 42 personality traits (also seen in Table below), which were suggested for usage as an aid for comparing brand personalities throughout different categories. Second, the authors assembled synonyms to the five personality traits with the online version of Encyclopedia Britannica‟s dictionary thesaurus function (Thesaurus, 2012), which according to Hamilton (2003) gives the best search results, because of its variety of information sources. However, the authors did not completely rely on the thesaurus. Some synonyms, provided by thesaurus, such as „hairy‟ and „amateur‟ could not be used, because they were not useful for our purpose and may misrepresent the findings. In addition, the thesaurus suggested words related to the brand personality dimension synonyms. These were used to enrich the dictionary known as “The dictionary of brand personality dimensions and brand personality synonyms” (See Appendix 5). The intention was mainly to find original synonyms for traits presented in the table below and for the five basic dimensions (sincerity, excitement, competence, sophistication and ruggedness), to enrich and enhance our data collection instrument/dictionary tool, to enhance content validity (Carmines et al, 1979)

3.6.2 Step

Three:

Designing

the

dictionary

Two independent reviewers contributed by collecting synonyms to the dictionary (Kolbe et al, 1991). A male masters student, who is fluent in nine languages (English, Swedish, Macedonian, Bosnian, Bulgarian, Croatian, Spanish, Russian and German), and a female masters student, who is fluent in five languages (English, Swedish, German, Albanian and French). The authors assumed that the language background of the reviewers would contribute to the study, due to how people with different linguistic backgrounds give meaning to words.

The authors enhanced the intercoder reliability, by bringing the two reviewers together to argue and justify their choice of words. Accordingly, there is a biasing effect of coder knowledge of variables extraneous to the content analysis, Banerjee et al., (1999). Consequently, to avoid that the referees tried to give what they thought the authors wanted they were not made aware of the research questions of the study. Lastly, the synonyms were included and implemented in the dictionary. To enhance the dictionary, a third reviewer -a lecturer from the department school of engineering- was included. The coder reviewed the dictionary independently and provided comments and suggested synonyms, which were also included in the dictionary.

WordStat was optimal software to use as it scanned for duplication of words to make sure that words did not overlap and was placed in two categories. Hence, the authors are confident the potential problem was prevented due to our selected software. In argumentative circumstances, the software has its own dictionary and „helped‟ the authors by asking the authors for permission to place potentially ambiguous words in other categories. Hence, there were only a few words that required investigation interference to decide and place words. In addition, some words such as “business-related”, “education”, “corporate”, “organization”, “business”, “company”, “learning”, “multinational” and “technology”, “just”, “open”, “nice”, “usual”, “very”, “well” were frequently used on all business schools‟ websites. Consequently, the words were eliminated because they were found to be „too ambiguous‟, „too common‟ or „frequently used‟

Along with stemming words, WordStat also excludes some words. The words excluded were: A, about, all, almost, also, always, an, any, are, as, at, be, because, been, being, both, but, by, can, could, do, every, feel, for, get, gets, got, has, he, her, him, his, how, if ,in, into, is, it, kind, like, lot, made, makes, me, more, most, much, my, myself, of, on, one, or, other, our, really, she, so than, that, the, there, these, they, this, to, too, type, us, very, was, were, what, when, where, which, who, will ,with, within, you. Consequently, the authors decided to permanently eliminate them from the gathered dictionary in order to develop a final list of 1241 words.

3.7 Validity & Reliability

3.7.1 Validity

Two criteria when conducting research are validity and reliability. These are used to be able to give the research credibility and are widely used. Reliability concerns the potential of the research could be done at another time by other with the similar prerequisites. This gives credibility and shows that, for example, the sample represents the group intended to study. Furthermore, if the research is too different from the first occasion, reliability cannot be met (Bryman et al, 2010).

quantitative research, validity is much difficult to get than in the qualitative research. This is mainly due to the channel between the researcher and respondent being longer and parts of the communication is lost along the way (Holmes & Solvang, 1997). There are different types of validity.

According to Yin (2008) internal validity is used when researchers are trying to determine if an event leads to another event, it is therefore in explanatory studies or causal studies. Yin (2008) mentions that the conclusion in a case-study drawn as a state cannot be observed directly. This means that problems can arise in case-study research as internal validity may lead to the researchers want to draw sound conclusion.

External validity is about concerns how well the results of a study can be generalized to other situations beyond the current case study, for example, if a result of a survey of changes in residential area is likely to apply in a different area (Yin 2008). Yins (2008) argue that the results can be generalized to a lagers population by comparing the case study with survey study.

Yin (2008) state that, when it comes to constructed validity it is important that researchers know in advance which specific factors that are going to be studied and then be able to see changes in them (Yin 2008). Yin (2008) discuss about two steps that must be ensured that the requirements have to be met as construct. The first requirement is that the changes that are measured must be selected and connected to the study´s original objectives. Second requirement is to show that the selected measure of the change reflects the specific kinds of changes that have been selected. There are numbers of avenues that can be used by scientists to increase construct validity of case study. Construct validity may increase by researchers using multiple empirical sources, and formulate an evidence trail for data collection. Another study for increasing this is by key informants review drafts from case study reports

articles‟ results favorable. We further tried to achieve validity by having a connection throughout the paper, with the focus on connecting the research to fit the purpose (Bryman et al, 2010). Miles and Huberman (1994) state that validity has for a long period, been a key issue of debate over the relevance of qualitative research. This study is conducted by using a qualitative, random data collection. The data is analyzed from words in different, existing MBA business schools websites in Europe, which means that secondary data was used. Further, the data was collected within five days, this because we wanted to minimize the chance for the websites to change. By collecting similar data from different websites the paper is considered as valid, in our opinion. The generalization of this paper can be seen as decent, this because Appendix nine shows that this study is generalizable with over 80 percent. Further, it is important to notice that the study is based on 10 business schools and that can be seen as to small group of sampling to generalize the paper for all the European school. However, the schools are similar and a representation of many others, which may enhance the degree of generalization of the results. Aakers (1997) brand personality model has been well accepted by others and that‟s one of the big reasons why we decided to implement it. The data information that has been used comes from reliable sources.

3.7.2 Reliability

The writers wanted to achieve reliability by being as consistent as needed in the research. To show reliability it is of importance to show how the data have been collected. Reliability is a measure of the extent to which study results are the same if it were to take place again. The goal of this is that if a similar case study was to be made again and using the same approach as previous researchers and thus reaches the same result as this. In this way, researchers on the results of a paper are stable and reliable. The Reliability is to reduce the error in a data. This is done by previous researchers‟ approach is used as carefully as possible (Yin 2008), (Bryman and Bell 2005).

study was made in another time we believe that the result will be the same as this time. We have explained the steps in a good way and by following our methodology a same study can with the same result will be made by others.

4. Empirical framework

The empirical chapter focuses on the presentation of data collected from the ten sampled European Business schools and their MBA programs’ websites, they will be compared within their Clusters before being combined later in the chapter. The first section will present the distribution of brand personality dimensions over the various web sites in tabular forms, with short explanations in words. Second, to make it easier for the reader to understand the gathered data, the authors will present a two dimensional correspondence map. Finally, there will be a precise presentation of how brand dimensions are connected to the ten schools websites. Note that most of the gathered data figure and tables are presented in the Appendix chapter.

4.1 Distribution of frequency and brand

dimensions

Table 4.1.1Cross-Tabulation of the Categorical Data and Aggregation of the Number of Words Identified by the Analysis Software, Cluster one

BPS Dimension/

Name of School Competence Sincerity Excitement Sophistication Ruggedness

Column Total HEC Paris 109 83 87 57 30 366 IE Business School 102 102 100 77 18 399 IMD Business school 106 80 78 32 32 328 London Business School 142 105 130 85 35 497 Oxford: Said 149 140 103 99 42 533 Row Total 608 510 498 350 157 2123 χ2 103.021 df 16 p< 0.001

Announcement For demonstration purposes the authors have rearranged rows and columns shown in the table. What

appears here as rows were earlier columns, and vice versa, when analyzed in WordStat.

The chi-square test (x2:103,021; df:16;p:0.001) shows that the row (i.e., the brand personality dimensions) and the column (i.e., websites) variables are related. The dimension-reporting results from Table 4.1.1 identify a statistically significant association between Aaker‟s five dimensions of brand personality and the MBA programs. In accordance to the P-value, one can clarify that there is a 99.9 percent assurance (relevance) that the relation between the personality traits and the schools are statistically ensured and not aleatoric.

Table 4.1.2: Cross-Tabulation of the Categorical Data and Aggregation of the Number of

Words Identified by the Analysis Software, Cluster two

BPS Dimension/ Name of School

Competence Sincerity Excitement Sophistication Ruggedness Column

Total HEC Genève 40 31 36 16 16 139 Kingston University 167 107 122 72 55 523 Paris School of Business 69 51 52 50 13 235 Sabanci Univeristy 199 149 98 47 46 539 Trinity College Dublin 188 215 123 72 58 656 Row Total 663 553 431 257 188 2092 χ2 527,7 df 16 p< 0.000

Announcement For demonstration purposes the authors have rearranged rows and columns shown in the table.

What appears here as rows were earlier columns, and vice versa, when analyzed in WordStat.

the brand personality dimension portrayed most in all programs in Cluster two, apart from Trinity College Dublin, was “competence”. “Sincerity” is the second most portrayed personality trait in almost all programs, while “ruggedness” represented the least portrayed brand personality, with 188 terms, which is consistent with the data in Table 4.1.2.

A chi-square test (χ2=527,7) in Table 4.1.3 shows that the brand personality dimensions- and web sites variables are related and significant among Cluster two schools. The chi-square test is a standardized measure comparing actual cell frequencies to expected cell ones and is used by correspondence analysts to standardize the frequency values and form the basis of association (Hair et al., 1998). The degree of freedom (df=16) measures the difference between the number of covariance and the actual number of coefficients in the proposed model, were higher values indicates a better fit (Cooper and Schindler, 2003). However, no threshold level of acceptance is established but values ranging between two and five often indicate that the model needs improvement (Hair et al., 1998), which is not the case for the model above.

Table 4.1.3: The Chi-squared Analysis for the individual Clusters

Table 4.1.3 presents the chi square test, degree of freedom and the probability of each Cluster. The results in Table 4.1.3 show in all circumstances, in accordance to Cluster one and Cluster two, that the brand personality dimensions (i.e., the rows) and the websites (i.e. columns) are related, verifying to evidence of strong dependency in each Cluster.

4.1.1 Distribution of frequency of brand

personality dimensions for Cluster

one and two

Cluster one

five-brand personality traits. The figure also illustrates the average distribution of the key words in the five schools and how it is obtained by aggregating the number of words identified by the analysis software WordStat. It appears that „competence‟ (31,7%) is the primarily trait that is frequently used among the top European business schools ( Cluster one) followed by „sincerity‟ (26,4%); „excitement‟ (20,6%); „sophistication‟ (12,3%) and least frequent appeared is „ruggedness‟ (9%). For more detailed information of each scholar in Cluster two, see Appendix seven.

In addition, the five brand personalities traits are further identified in depth for each school, separately, in Appendix eight, were the words representing the personality are distinguished by colors. It appears that „competence‟ is the personality trait that is most frequently used among the schools in Cluster one (HEC Paris: 109; IE Business School: 102; IMD Business School: 106; London Business School 142; Oxford: 149) followed by „sincerity‟ (HEC Paris: 83; IE Business School: 102; IMD Business School: 80; London Business School: 105; Oxford: 140) and „excitement‟ (HEC Paris: 87; IE Business School: 100; IMD Business School: 78; London Business School 130; Oxford: 103). „Sophistication‟ is the third most frequently used personality dimension (HEC Paris: 57; IE Business School: 77; IMD Business School: 32; London Business School 85; Oxford: 99 whilst „ruggedness‟ (HEC Paris: 30; IE Business School: 18; IMD Business School: 32; London Business School 33; Oxford: 42) was the least frequent used dimension in the schools in Cluster one.

Cluster two

Moreover, the five brand personalities traits are further identified in depth for each school, separately, in Appendix eight, were the words representing the personality are distinguished by colors. It appears that „competence‟ (HEC Genève: 40; Kingston University: 167; Paris School of Business: 69; Sabanci University: 199; Trinity College: 188) is the personality trait that is most frequently used among the schools in Cluster two, followed by „sincerity‟ (HEC Genève: 31; Kingston University: 107; Paris School of Business: 51; Sabanci University: 149; Trinity College: 215). „Excitement‟ (HEC Genève: 36; Kingston University: 122; Paris School of Business: 52; Sabanci University: 98; Trinity College: 123) is frequently used among the schools and „sophistication‟ (HEC Genève: 16; Kingston University: 72; Paris School of Business: 50; Sabanci University: 47; Trinity College: 72) is less frequently used. „Ruggedness‟ (HEC Genève: 16; Kingston University: 55; Paris School of Business: 13; Sabanci University: 46; Trinity College: 58) is the least frequently used brand personality trait in Cluster two.

4.2 Two dimensional correspondence map

Figure 4.2.1: Correspondence analysis map of five European Business Schools‟ websites in

For verification purposes Figure 4.2.1 is conducted to crosscheck the findings and re-examine earlier presented arguments from previous discussion. The figure lists the top five ranked Business schools ranked by Financial Times in Cluster one. The figure also presents the five brand personality dimensions. The two-way dimensional correspondence map accounts for more than 92 percent of the variance, with eigenvalues of 64,928 and 27,804. In terms of reliability, the total variance of greater than 70% is significant because of the exploratory nature of the study (Hair et al., 1998). The correspondence is significant up to at least 92 percent and should provide an „accurate‟ description of the data. For more detailed information about variables and variables coordinates of Cluster one, see Appendix nine

Figure 4.2.2: Correspondence analysis map of five European Business Schools‟ websites in