THE EFFECT OF SPORTS SPONSORSHIP ON BRAND ATTACHMENT

THE EFFECT OF SPORTS SPONSORSHIP ON BRAND ATTACHMENT

SPONSORING BRANDS

SPONSORING BRANDS

The Effect of Sports Sponsoring on Brand Attachment

Author: Laura Wilpstra

Department: Department of Economics and Business Qualification: Master Thesis

Completion Date: August 30, 2011

Address: Helper Oostsingel 31a, 9722 AR, Groningen Tel #: 06-53232506

E-mail: laurawilpstra@mac.com

MANAGEMENT SUMMARY

The marketing phenomenon sports sponsorship is a focal concept in this study; The effect of this concept on brand equity is investigated.

Until now the effect of sports sponsorship has mostly been investigated on the lower levels of brand equity (as in the pyramid defined by Keller 2008). In this study the effect on a higher level, brand attachment, is investigated. It is interesting to see if sports sponsorship can also have an influence on brand attachment, because it is important for a brand that consumers are attached to it.

An attached consumer will e.g. not switch brands easily.

The literature review identifies several variables that can influence the proposed relationship between sports sponsorship and brand attachment, and leads to a conceptual model. Consumer connection, which is the extend to which the consumer feels emotionally involved with and attached to a sponsored object, is identified as a mediator. It is thought that consumers can feel connected/attached to a sponsored object (e.g. event or individual) and that this feeling will have a spillover effect on the brand (the consumer will feel more connected to the brand).

This consumer connection is presumed to be influenced by the type of sponsored object (event or individual in this study).

Furthermore, it is likely that if a consumer feels connected to the sponsored object that he/she will be thankful to the brand for the sponsorship. This thankfulness is presumed to have a moderating effect on the relationship between consumer connection and brand attachment. Moreover, the perceived intention (goodwill or commercial) of the sponsorship is presumed to influence thankfulness.

An experiment is carried out in order to collect the data that are necessary to test the formulated hypotheses. There are four experimental groups (individual vs event and goodwill vs commercial) and one control group.

The conducted research proves that there is an effect of sports sponsorship on brand attachment. This effect is not direct but is fully mediated by the consumer connection with the sponsored object. The higher the consumer connection is, the higher the brand attachment will be. Furthermore, the consumer connection proves to be higher in the case that the sponsored object is an event. Moreover, thankfulness proves to have a moderating effect on the relationship between consumer connection and brand attachment. The mediation effect of consumer connection on brand attachment is stronger when the level of thankfulness increases. The level of thankfulness is positively influenced by consumer connection.

The results imply that indeed there is a spillover effect between consumer

connection and brand attachment. Furthermore, sponsoring an event seems most

beneficial since this significantly increases consumer connection. However, this

might be different in a different context. Hence, it is recommended to investigate

PREFACE

This report is the final piece of work that I will hand in at the University of Groningen. Handing in this report means that I have almost finished my Master of Science in Business Administration. It also means that a different life will start, a working life, which will hopefully be as exiting as my student life was/is.

Writing this master thesis has challenged me academically and required to put all of what I have learned over the years into one single report. I particularly enjoyed the statistical analysis part of the thesis, since I had to conduct analyses that were completely new to me and I enjoy to learn. I think that my academic background is now sufficient, that I am ready to start putting into practice what I have learned, and ready to learn from practice.

Whilst writing this thesis I have had much support from my supervisor Mr Alsem.

I would like to thank him for sharing his witty comments and interesting ideas.

Furthermore, many thanks goes to my second supervisor Y.C. Ou who has provided me with some very useful comments to improve my thesis. Moreover, I would like to thank my parents for their mental and financial support, without them I would not be in the position to study in the first place.

I hope that you will enjoy reading my final report.

Laura Wilpstra

TABLE of CONTENTS

MANAGEMENT SUMMARY PREFACE

1. INTRODUCTION

1.1 Theoretical Background 2 1.2 Problem Statement 3 1.3 Paper Structure 3 2. LITERATURE REVIEW

2.1 Sports Sponsorship 4 2.2 Brand Equity 10

2.3 Conclusion of Literature Review 12

2.4 Effect of Sponsorship on Brand Attachment 13 2.5 Conceptual Model 14

3. RESEARCH DESIGN 3.1 Design 15

3.2 Manipulation of the Variables 16 3.3 Subjects 16

3.4 Procedure 17

3.5 Testing of the Variables 18 4. RESULTS & ANALYSIS

4.1 Sample Impression 20 4.2 Scale Reliability 21 4.3 Descriptive Results 23 4.4 Hypotheses Testing 24

5. CONCLUSION and DISCUSSION 5.1 Conclusion 33

5.2 Discussion 34

5.3 Managerial Implications 35

5.4 Limitations and Future Research 36

REFERENCES

1. INTRODUCTION

In this introductory chapter a theoretical background to the research is presented, followed by the problem statement, research questions and an overview of the structure of the paper.

1.1 Theoretical Background

Sponsorship is a relatively new phenomenon in marketing communication. It is the latest edition to the promotion mix (Lagae, 2006). Sponsorship is nowadays a well established part of the communication mix.

A firm can sponsor many different things, e.g. sports, arts and good causes. In this research the focus will be on sports, because sports bond people from every layer of society. “Sport is not just a convenient or casual way of filling in leisure time, but is also an important cultural institution that delivers significant social benefits to a diverse spread of communities” (Smith & Stewart, 2009). Furthermore, sports sponsorship creates the highest level of awareness compared to culture sponsorships, sponsorship of social aid organisations and sponsorship of TV programmes (Hansen et al 2006).

Sponsorships usually involve significant amounts of money. Hence, sponsors demand more and more in return for their sponsorship. It is important for marketers to justify the choice for sponsorship to the boardroom. Proving that the sponsorship enhances brand equity is therefore important, since brand equity is of great value to the firm and can have a strong, firm specific, risk-reducing effect.

(Rego, Billett & Morgan, 2009). Amis, Slack and Berrett (1999) argue that if a firm sees sponsorship as a valuable source and invests time and effort, than be a sustainable competitive advantage.

The brand equity pyramid as defined by Keller (2008) has four layers. The lower two levels concern brand awareness and brand image. The upper two levels concern the feelings around a brand, brand loyalty and brand attachment. In previous research, sponsorship has mostly been related to the lower levels in the pyramid of brand equity, namely brand image and brand awareness. However, previous research by Madrigal (2000) and Thompson (2006) proves that consumers can become very much emotionally involved and attached to e.g. famous athletes or soccer team, they can become part of the consumer’s self. Furthermore, research has proven that there can be a spillover effect from the sponsored object to the brand (e.g. Crimmins & Horn 1996, Gwinner & Eston 1999, Martensen et al 2007).

Hence, it is interesting to see if sponsorship can also have an influence on a higher level of brand equity, namely brand attachment. Brand attachment is defined by Park et al (2010) as “the strength of the bond connecting the brand with the self”.

It is assumed that if the consumer sees e.g. a soccer team as part of the self, this

can have a spillover effect on the brand.

A study by Sirgy et al (2008), proves that the level of identification consumers have with the people that visit a sponsored event can have a spill over effect on the brand and that it can enhance brand equity if the consumer feels a high level of identification, because it enhances behavioural brand loyalty. Behavioural brand loyalty refers to how often consumers purchase a brand and how much they purchase (Keller, 2008).

Brand attachment is a step further than behavioural brand loyalty (Keller, 2008) and hence this study will add to previous research by investigating whether sports sponsorship can also have an effect on an even higher level of brand equity.

Furthermore, other variables will be considered than in previous studies. It is possible that the relationship between sports sponsorship and brand attachment is e.g. mediated or moderated by certain variables. These variables come from the literature review that is presented in chapter 2.

1.2 Problem Statement

To what extend does sports sponsorship influence brand attachment, does the type of sponsorship matter and is the influence direct or are there mediating/

moderating variables involved?

1.3 Paper Structure

Following this introductory chapter is a critical review of the relevant literature in

the field of (sports) sponsorship, the effect sponsorship has on brand equity and the

concept brand equity with a focus on brand attachment. Secondly, the research

design for this study is explained. Followed by an overview and analysis of the

results of the research. Finally, a discussion of the results is presented and

recommendations for future research are given.

2. LITERATURE REVIEW

2.1 Sports Sponsorship

Sponsorship can be defined as “a business relationship between a provider of funds, resources or services and an individual, event or organisation which offers in return some rights and association that may be used for commercial advantage” (Sleight 1989). The firm thus receives something in return for their sponsoring, which makes it fundamentally different from charity.

2.1.1 Forms of sports sponsorship

There are different forms of sports sponsorship that can be chosen by a firm (Lagae 2006). A firm can choose to sponsor an event, a competition, a sports team, an individual sports person, a sports federation or a sports accommodation. The decision for which form of sports sponsorship is the best choice for the firm should be based on factors concerning the characteristics of the sports form, e.g. media exposure, but also on factors concerning the response of the consumer, e.g.

consumer connection. Furthermore, the fit between the sponsored object and the brand is proven to be essential. In the next two paragraphs the factors concerning these two topics are addressed.

2.1.2 Characteristics of the sports sponsorship forms

Each form of sports sponsorship has different characteristics and these characteristics are important to consider when choosing a sponsoring object (Lagae 2006).

Making the consumer aware of the sponsorship is understandably an important point, because if the consumer is unaware of the sponsorship it will never benefit the firm. Besides marketing communications by the firm, attention from the media can aid the creation of sponsorship awareness. Hence, it is important to consider how much media attention a sponsorship form will receive, since not every form will receive the same deal of media attention.

Since marketing communications are also important for creating the sponsorship awareness, it should be possible to integrate the sponsorship in the marketing communications of the firm. For example a supermarket will be more interested in a sponsorship form with a wide target group, because it will enable them to combine promotions in the supermarket with the sponsorship in order to create awareness. Furthermore, Quester and Thompson (2001) argue that leveraging the sponsorship with further marketing communications is essential to its success, because it results in more positive attitudes toward the sponsor and raises the chance that the consumer will interact with the sponsor in the future.

Another important factor to consider is the amount of sponsors that are involved

with the sponsored-object. Some sponsoring forms have just one single sponsors

where others have multiple sponsors. According to Ruth and Simonin (2006)

sponsorships can be perceived as goodwill oriented and sales oriented. In general

This is especially the case when there is only one sponsor involved because the sponsorship is than perceived as a distinctive action. However, when more sponsors are involved it makes the sponsorship less distinctive and hence the evaluation of the sponsorship will also be less extreme. The effect of goodwill oriented sponsorship is than less positive and the effect of the sales oriented sponsorship less negative.

In general it is most likely more beneficial to the firm if the sponsorship is perceived by the consumer as goodwill oriented rather than as sales oriented.

However, in both cases it is thus import to consider the amount of sponsors that are involved in an event. Moreover, a firm should consider the image of the other sponsors since it is possible that this will have a spill over effect on their own brand if the firm is involved in the sponsoring of the same event.

In case of the event form of sponsorship it is important to consider the size of the event. Akaoui (2007) reports that sponsoring a major event is more influential than a smaller event. This seems logical if the goal is to create brand awareness, since more consumers are reached with a major event than with a smaller one. However, brand awareness, even though it is highly important, is not always the only goal of sponsorship. Hence, smaller events might be a better choice for the firm depending on the goal and other factors that influence the response of the consumer, which are discussed in the next paragraph.

In conclusion, important characteristics of sponsorship forms that the firm should take into consideration are:

‣ The amount of media attention that the sponsorship form receives

‣ The possibility to integrate the sponsorship into the marketing communications of the firm

‣ The amount of sponsors that are involved with the sponsored object

‣ The size of the event

2.1.3 Factors influencing the response of the consumer

There are several factors identified in previous research that influence the response of the consumer to sponsorship. In the following paragraphs literature regarding the influence of the sponsored object and the influence of the sponsor itself is discussed. Many studies consider events as the sponsorship form. An event is an experience to the consumer and they can become much involved. However, the consumer can also become involved with e.g. a single athlete or a team.

Furthermore, an experience can be created around those forms as well. Therefore, other forms of sports sponsorship are also part of this study and not only events.

Consumer connection

Research by Madrigal (2000) proves that people can become very much emotionally

involved with e.g. their favourite athlete or sports team. An athlete or sports team

can become part of the person’s self. He or she is than likely to behave

prototypically because the ‘group’s’ norms, values and goals are accepted .

This affective commitment to a sponsored object has also the focus in research by Lings and Owen (2007). The authors argue that the longer the association with the team, the higher the perceived prestige of the team and higher the level of participation with the team, the higher the affective commitment with the team.

Furthermore, they argue that the higher the affective commitment is, the more likely it is that the consumer will identify with the team and adopt its norms and values. The greater the identification with the team the more likely the consumer is to purchase the sponsor’s products. If the team is winning this effect is significantly larger than when the team is losing.

Additionally, research by Thomson (2006) proves that consumers form strong attachments to ‘human brands’, e.g. famous athletes. The author states that the consumer uses words like love and adoration to describe the attachment that they feel to a ‘human brand’. This indicates that a consumer can feel as strongly about a

‘human brand’ as he or she can feel about e.g. a family member.

The consumer firstly needs to be attracted to the ‘human brand’, because otherwise it is unlikely that the consumer will become attached. Furthermore, the author states that the fulfilment of two fundamental human needs are at the basis of the formation of this attachment:

‣ Autonomy; a person’s need to feel that his or her activities are self- chosen, self-governed and self-endorsed (Deci & Ryan 2000)

‣ Relatedness; a person’s need to feel a sense of closeness to others (Deci

& Ryan 2000)

Feelings of autonomy are found to be created by ‘human brands’ that make the consumer feel appreciated, empowered and understood. If a ‘human brand’

promotes acceptance, openness and belonging, than feelings of relatedness are created. In order to enhance the attachment it is important that there is interaction between the consumer and the ‘human brand’. Both the quality and quantity of this interaction are of significant importance. Furthermore, the ‘human brand’ should not be perceived as performing and act and being unauthentic, because the consumer might feel manipulated.

Furthermore, research by Meenaghan (2001) proves that highly involved fans are most sensitive to the sponsorship activities. If they are satisfied by the sponsor’s behaviour towards the sponsored object they are the most likely to favour the sponsor. However, the reverse is also true, if they do not like the behaviour of the sponsor the highly involved fans are also the most likely to turn against the sponsor.

This attachment to ‘human brands’ makes it therefore important to select a

sponsored object that is well liked by the target group, because often ‘fans’ of the

sports club or event it self will express their ‘gratitude’ for helping their club,

toward the sponsor (Speed & Thompson 2000, Meenaghan 2001). Furthermore, a

positive attitude toward the event is associated with a positive response toward the

sponsor. They will thus respond more positively toward the sponsorship.

Moreover, research by Martensen et al (2007) indicates that the emotions that an event calls will have a spillover effect on the brand. Negative emotions have a stronger effect than positive ones. Hence it is of great importance to make sure that the event will call positive rather than negative emotions among the target audience.

These emotions that are called by an event have an influence on the event attitude that the consumer holds and the authors prove that this event attitude has a strong positive influence on the brand attitude. Additionally, Meenaghan (2001) argues that the favourable attitudes toward a sponsorship are more pronounced when they are emotionally involved with the sponsored object.

This is supported by Martensen et al (2007) who argue that the more involved the consumer is with the event the more positive and the less negative negative emotions are called. The authors also argue that the more involved the consumers are with an event the more the attitude toward the event increases. Furthermore, the authors argue that a consumer who is familiar with an event has a more varied frame of interpretation which also influence event attitude.

In addition, research by Sirgy et al (2008) proves that customers of a particular product who can identify with people attending an event that is sponsored are likely to feel more loyal toward the sponsor, because the level of identification with the event has a spill over effect on the brand. This effect is moderated by the level of awareness of the sponsorship that the customer has and the level of involvement with the event. The authors suggest that a firm should make use of a supporting marketing campaign to enhance the awareness and involvement of the consumer with the sponsored sports event.

Sponsor- sponsored object fit

Speed and Thompson (2000) argue that in order to make the sponsorship effective, the consumer should be able to identify a fit between the event and the sponsor. If there is a perceived fit between the sponsor and the event, it is more likely that the sponsor will correctly be identified and recognised (Venkataramani, Pham &

Wakefield 2006). However, the authors also warn that it is not always a correct identification but sometimes also a educated plausibility guess, by the consumer.

Furthermore, a fit creates the conditions for a value transfer from the event to the brand (Martensen et al 2007). The authors argue that a perceived fit between the brand and the event has an effect on the consumers responsiveness to the message and the emotions that an event calls.

Moreover, Gwinner and Eston (1999) argue that the image of the event should match with the brand image, because the image can transfer between the event and the brand.

In the case that there is not an immediate fit, a sponsoring firm can enhance their

association to the event by investing in marketing communications around the

event (Venkataramani, Pham & Wakefield 2006). Furthermore, marketing

communications can be used to send a message to the consumer that creates a

Sponsor image

According to Speed and Thompson (2000) the response of the consumer is highly influenced by the opinion and image they have of the sponsor prior to the event.

The response of the consumer will be more positively if he or she has a positive attitude toward the sponsor prior to the event, because it is approached more positively.

Furthermore, if a sponsor is perceived to be sincere the sponsorship is not just seen as another form of commercial activity and the consumer will be more sensible to the philanthropic dimension that sponsorship has. The more sincere the sponsor is perceived by the consumer the more positively is the response of the consumer (Speed & Thompson 2000). Research by Meenaghan (2001) supports this finding by proving that sponsorship results in lower defence mechanisms compared to advertising.

Meenaghan (2001) argues that the consumers beliefs that commercial sponsorship directly benefits e.g. sports and that the consumer therefore exhibits a certain goodwill factor towards sponsors. However, sports sponsorship generally receives lower levels of goodwill than sponsorship categories that are less exploited.

Furthermore, sponsorship is seen by the consumer as more subtle and indirect compared to advertising, because it is moderated by the involvement of e.g. sports in the communication. Sponsorship takes place in the background of the activity in which the communication takes place. This subtleness makes the consumer perceive sponsorship as less forceful and also arouses less anxiety than advertising.

The moderating effect of the sponsorship activity and the subtleness of sponsorship lead to a lower state of mental alertness and therefore influences the defence mechanisms. Additionally the different delivery context of the sponsorship message compared to advertising also is of influence. Moreover, the halo of goodwill around sponsorship makes the consumer more receptive. Although, the results of the the study by Meenaghan (2001) clearly state that the consumer perceives sponsorship more favourably to advertising, the effectiveness was not considered in the study and it can therefore not be stated that sponsorship is more effective than advertising.

A sponsoring firm can receive superior benefits from its sponsorship if the

consumer has a positive image of the firm and perceives the sponsorship to be

sincere. However, research by Speed & Thompson (2000) shows that a firm should

not just sponsor anything and everything, because it can lead to an increase in the

perceived level of ubiquity. If the consumer believes the firm is just sponsoring

anything and everything, it will lead to a reduction in response to all sponsorships

of the firm. This might be true in some cases however one can also argue that the

more a firm sponsors, the more visibility and attention the firm receives. Hence,

the more brand awareness is created.

2.1.4 Spillover effect between the sponsored object and the brand As discussed previously in this chapter it is likely that there is a spillover effect between the sponsored object and the brand.

Gwinner and Eston (1999) argue that it is important to see what the target consumers’ perception is of the image of the event. It is important that this is consistent with the image that the firm wants for its brand because the image of the event can transfer onto the brand. Altering marketing communications around the brand prior to the event can enhance this image transfer.

Furthermore, previously discussed research by Martensen et al (2007) proves that event attitude has an influence on brand attitude. This attitude is influence by emotions and event emotions have an effect on brand emotions.

Additionally, Crimmins and Horn (1996) argue that, through sponsorship, a link is created in the consumer’s mind between the brand and an event or organisation, that is already highly valued by the consumer. This link plays an important role in the persuasive impact that sponsorship has. The strength of the link, duration of the link, gratitude felt due to the link and the perceptual change due to the link form the combination that determine the persuasive impact.

2.1.5 Effect of sponsorship on brand equity

The vast majority of previous research on effects of sponsorship has focussed on the lower levels of brand equity, brand awareness and brand image, as defined in the brand equity pyramid by Keller (2008). However, there is reason to believe that sponsorship can also influence a higher level of brand equity; Brand attachment.

First of all, research by Lings and Owen (2007) proves that sports sponsorship can have an effect on a higher level of brand equity, namely purchase intentions.

Secondly, Sirgy et al (2008) prove that sports sponsorship can also have an effect on brand loyalty, which is at the same level as brand attachment.

Furthermore, previous research proves that there can be a spillover effect between the sponsored object and the brand (e.g. Gwinner & Eston 1999, Martensen et al 2007). Additionally, previous research also proves that the consumer can become attached to a sponsored-object and that a connection can exist between the sponsored-object and the self (Thompson 2006). Research by Madrigal (2000) proves that a benefit derived from sponsorship is the opportunity for the firm to link itself to something that is part of the consumer’s extended self. Therefore, it is presumed that this attachment to the sponsored object can also have a spillover on the brand. Hence, sponsorship is presumed to have an effect on brand attachment.

In the next paragraph the concept of brand equity and particularly brand

attachment is discussed into more detail. This is followed by a paragraph that goes

deeper into the presumed relationship between sponsorship and brand attachment.

2.2 Brand Equity

2.2.1 Models

Many have conceptualised brand equity, two of the best known conceptualisations are of Aaker (1991) and Keller (1993). There are similarities and differences between the models.

Keller specifically uses the term Customer-Based Brand Equity, which is defined as “the differential effect of brand knowledge on consumer response to the marketing of the brand (p. 2). The brand knowledge that a consumer has, consists of the brand awareness and the brand image, and influences the reaction the consumer has to an element of the marketing mix of a particular brand compared to another. High levels of brand awareness and a positive brand image (consists of brand associations) can increase the probability of brand choice and can enable the firm to demand premium prices for example.

Consistent with Keller (1993) the model of Aaker (1991) has brand awareness and brand associations as dimensions of brand equity. However, Aaker’s model consists of two additional dimensions namely; Brand loyalty and Perceived Quality.

Furthermore, Aaker argues that the consumer also receives value from these brand equity dimensions. The consumer will feel more confident about the purchase decision, enhance their processing of information and enhance their satisfaction with the use of the product. The value brand equity will have to the firm is consistent with the findings of Keller (1993).

The models of Aaker (1991) and Keller (1993) both do not take the more emotional side of brands into consideration. Park et al (2010) argue that brand attitude strength and brand attachment are two critical drivers of brand equity. The study shows that brand attachment is a better predictor of actual behavior than brand attitude strength. Park et al (2010) argue that the more attached consumers are to a brand, the more willing they are to invest time, money, energy and reputation to maintain or deepen their relationship with the brand.

2.2.2 Dimensions of brand equity in this study

In the paragraph 2.1.7 it is discussed that prior research on sponsorship has mainly focussed on the lower levels of brand equity and there is a presumed relationship between sponsorship and a higher level of brand equity; brand attachment. Hence, the dimension brand attachment will be the focal dimension in this study.

A second dimension that is considered is brand awareness, because a brand node or a trace of memory about the brand is needed in the memory of the consumer, since this determines how easily different types of information become attached to this brand in memory (Aaker 1991, Keller 1993, Rossiter & Percy 1987).

2.2.3 Brand awareness

“Brand awareness refers to the strength of the presence of the brand in the memory of a consumer” (Aaker 1991). Brand awareness also has two dimensions;

brand recognition and brand recall (Keller 1993).

If consumers are asked to e.g. name five cheese brands, they have to retrieve the brand from their memory. This is a situation that relates to brand recall: “The consumers ability to retrieve the brand when given the product category, the needs fulfilled by the product category, or some other type of probe as a cue” (Keller 1993). Consumers can also be given the brand as a cue and if they recognise it and remember to be previously exposed to it, than one can speak of brand recognition (Keller 1993).

2.2.4 Brand Attachment

Attachment is defined as an emotion-laden, target-specific relationship between a person and a target (Bowlby 1979). It is a fundamental human need. Bowlby (1969, 1980) was the first to start developing an attachment theory. The author argues that the interaction in early childhood between the child and the caregiver forms the foundation for how other relationships are formed in later life. The relationship between child and caregiver determines how a person sees the self and others. A person can have a positive or a negative view. Many other researchers have build upon this theory, finding that other relationships e.g. between romantic partners and friendships, are guided by the principles of the attachment theory (Hazan &

Shaver 1987, Ainsworth 1989, Trinke & Bartholomew 1997).

More recently, attachment has also been introduced to marketing research, primarily in the field of consumers becoming attached to a brand. One of the most recent studies that particularly views brand attachment as a driver of brand equity is a study by Park et al (2010). This study is used here to define and explain brand attachment.

Park et al (2010) define brand attachment as “the strength of the bond connecting the brand with the self”. It is mental representation that involves feelings and thoughts the brand and the brand’s relationship to the self. Brand attachment has two dimensions; Brand-self connection and Brand Prominence. The dimensions are both important, because brand-self connection explains the bond between the brand and the self, and brand prominence explains the strength of the bond.

Brand-self connection reflects the cognitive and emotional bonding between the brand and the self. The consumer can feel connected to the brand, because it e.g.

represents who they are or is in line with their goals. The consumer develops a oneness with the brand by establishing cognitive links between the brand and the self, that are inherently based on emotions and involve feelings about the brand e.g. happiness or sadness.

Brand prominence is defined as “the salience of the cognitive and affective bond

that connect the brand to the self” by Park et al (2010). The feelings and thoughts

that a consumer has about a brand, develop over time and become part of their

memory. The thoughts and feelings about different brands can differ in the

easiness and frequency that they pop into a consumers mind. This can make a

great difference for the brand attachment that a consumer has. The more easily

and frequent brand-related thoughts and feelings are brought to mind, the greater

is the brand attachment if the brand-self connection is the same.

Brand attachment has a significant effect on the actual purchase behaviour of consumers. The higher the brand attachment the more likely they are to actually purchase it. Furthermore, consumers who have a high Brand attachment are also less likely to purchase competing brands. Moreover, a consumer with high brand attachment is also less likely to purchase any substitutable alternative products in categories that satisfy the same need.

A consumer with high brand attachment is more willing to invest time, money, energy and reputation. This makes the differential effect in the consumer’s response to the marketing of the brand. The brand can e.g. demand premium prices and the consumer will go through more trouble to get the product, which makes the distribution less of an obstacle.

2.3 Conclusion of Literature Review

There are many forms of sports sponsorship and a firm needs to consider all the factors that are mentioned previously into consideration when making a decision about a sponsorship deal. Furthermore, there are many factors that can influence the response of the consumer toward the sponsor. Consumers can be very much involved with and attached to the sponsored object, which is of great influence.

Furthermore, the perceived fit between the sponsor and the sponsored object is proven to play an important role and also the image that a consumer has of the sponsor and the sponsorship.

The response of the consumer and the effects on the sponsoring firm have been the subject of many studies. The effect on brand equity is of great importance because it is of great value to the firm. As already discussed sponsorship has primarily been related to lower levels of brand equity in previous research. However, since higher levels of brand equity are more beneficial to the firm and because there is reason to believe that sponsorship can have a positive influence on these higher levels, the focus in this study is on a higher level namely brand attachment. In order to enable consumer to become attached to a brand, brand awareness is necessary.

Furthermore, it is necessary to create sponsorship awareness to enable consumers to respond to the sponsorship. Hence, a marketing campaign surrounding the sponsorship is necessary, it will create both brand and sponsorship awareness.

Moreover, sports sponsorship itself will of course create brand awareness.

2.4 Effect of Sponsorship on Brand Attachment

Consumers can become significantly attached to e.g. a sports team. This attachment toward the team calls many emotions on the consumer. They will feel affection towards the team and can identify with norms and values. The team is part of their ‘self’, the consumer can relate the team to their self-image. This is referred to as consumer connection in this study. These emotions are presumed to have a spillover effect on the sponsor of e.g. a sports team. Hence, the consumer will also feel as if the sponsor is part of their ‘self’. Since brand attachment is defined in this study as “the strength of the bond connecting the brand with the self” (Park et al 2010), sponsorship is perceived to have an effect on brand attachment.

H1: Sponsorship has a positive effect on brand attachment

However, it is also presumed that this effect is (partially) due to the involvement and attachment a consumer has with the sponsored object; Consumer connection.

Therefore, consumer connection is presumed to be a mediator in the relationship between the sponsorship and brand attachment. It is presumed that this consumer connection is greater in case of an individual athlete, because it involves a human being with emotions.

H2: Consumer Connection positively mediates the effect of sponsorship on brand attachment

H3: Consumer Connection is greater in case of an individual athlete than in case of an event

According to Ruth and Simonin (2006) sponsorship perceived as goodwill is more beneficial to the firm than a sponsorship perceived as commercial, because it calls more positive emotions with the consumer. This perceived intention is likely to also have an effect on brand attachment; The positive emotions are likely to have a positive effect on brand attachment, at least more positive than when the sponsorship is perceived as commercial.

H4: A sponsorship perceived as goodwill has a more positive effect on brand attachment than a sponsorship perceived as commercial

A sponsor can enable e.g. a sports team to perform well, because the team can benefit from the sponsored financial or other resources. Hence, a sponsor is of significant importance for the sponsored object. It is likely that consumers who are highly involved with e.g. the sports team also understand the importance of a sponsor for ‘their’ team. Therefore, it is presumed that they will feel thankful toward the sponsor for ‘supporting’ their team. Therefor the level of consumer connection is presumed to have an effect on thankfulness.

H5: Consumer connection has a positive influence on the thankfulness of the

consumer toward the sponsor

This thankfulness is presumed to create a feeling of the brand being on the ‘same side’ as the highly involved consumer. Both the brand and the consumer are supporting the team. This feeling is presumed to have an effect on the connection between the brand and the self of the consumer. Furthermore, consumer connection influences thankfulness (H4). It is therefore presumed that thankfulness will enhance the relationship between consumer connection and brand attachment.

H6: The mediation effect of consumer connection on brand attachment is positively moderated by thankfulness

Much research is dedicated to the fit between the sponsor and the sponsored object.

There are many studies that prove the impact of this fit on the effectiveness of sponsorship. In this study it is investigated if fit can also have a positive moderating effect on the relationship between sponsorship and brand attachment.

H7: Fit has a positive moderating effect on the relationship between sponsorship and brand attachment

2.5 Conceptual Model

Figure 2.1: Conceptual Model

3. RESEARCH DESIGN

In order to be able to investigate whether the proposed hypotheses can be excepted or not, experimental research is conducted. The variables as presented in the conceptual model need to be manipulated and tested. In this chapter an explanation of the phases of the experiment and the research subjects is given.

3.1 Design

Quantitative, experimental research is conducted to test the proposed hypotheses in chapter 2.

Sponsorship is the main independent variable and the choice is made to manipulate the sponsored object (individual vs event) and the perceived intention (goodwill vs commercial). This is done because an effect on the dependent variable brand attachment is expected. Furthermore, it is expected the manipulation will also have an effect on the level of consumer connection and the level of thankfulness. This is necessary because the variables cannot be manipulated themselves and differences in them are needed for the purpose of this study.

The design that is used is a factorial fully-crossed design with four experimental cells. The independent variables that are manipulated in the design are Event vs Individual and ‘Goodwill’ Sponsorship vs ‘Commercial’ Sponsorship. The dependent variable is Brand Attachment. In order to avoid test re-test issues and to get absolute evaluations a between-participants design is used. The participants can thus participate only in one experimental cell and are therefore not influenced by other experimental cells. Besides the experimental cells there is a control group to check the influence of the manipulations against a group of participants that is not manipulated.

Goodwill Commercial

Individual ^{2 3}

Event ^{4 5}

* Control group is group 1

Table 3.1: Experimental design

3.2 Manipulation of the Variables

The focal brand in the research is Land Rover and equestrian is the focal sport.

Four articles are written to manipulate the independent variables. The individual that is focal in the article for experimental cell 2 and 3 is Edward Gal, who is a famous equestrian athlete in The Netherlands. The event that is focal in experimental cell 4 and 5 is the European Championships Dressage which are held in Rotterdam (NL) this year.

The articles that represent the ‘Goodwill’ cells say how important it is for the athlete/event and it has quotes in which the thankfulness and happiness about the sponsorship is expressed. The articles that represent the ‘Commercial’ cells say something about the goal of Land Rover and about what the company hopes to get out of the sponsorship. Furthermore, something is e.g. stated about the board of directors that sits in the VIP lounge.

A pre-test determines that there is a perceived fit between the brand Land Rover and equestrian sports. Furthermore, the difference between ‘Goodwill’ and

‘Commercial’ is clear to the five participants in the pre-test. The participants of the pre-test each read all four articles and were asked to indicate the extend to which they agreed (7-point Likert scale) to the following statements for each of the articles:

‣ An event is sponsored by Land Rover

‣ An individual is sponsored by Land Rover

‣ Land Rover purely has commercial interest in the sponsorship

‣ Land Rover benefits greatly from the sponsorship

‣ The only reason that Land Rover sponsors is to increase sales

‣ The sponsorship is very beneficial to the event/individual

‣ Without the sponsorship the event/individual would not perform as well

‣ Land Rover sponsors to genuinely help the event/individual

‣ Land Rover has no benefits of the sponsorship

3.3 Subjects

A total of 125 consumers participate in the research. Each experimental consists of

25 participants and the control group also consists of 25 participants. The

participants are of various ages, no one was excluded. A pre-requisite is that the

participants have a connection with equestrian sports and in order to reach a large

population the questionnaire is posted on the forum www.bokt.nl. This is the

largest ‘horse forum’ in The Netherlands with over 200.000 members and it is

therefore a good source for participants for this research.

3.4 Procedure

A tool is needed to gather the data that are necessary to investigate the purposed hypotheses. The tool that is chosen here is a questionnaire with statements to which the participants can indicate to which extend they disagree/agree. A questionnaire is used to obtain the data from the participants in each of the experimental groups and the control group. There are five different questionnaires in total. In four questionnaires the different manipulations are used and in one questionnaire there is no sponsorship announcement present (control group).

The links to the questionnaires are posted in a special ‘topic’ on the forum. The links are posted in random order. The post clearly states that everyone should just fill out one questionnaire and the participants are thus randomly assigned to a group. The questionnaire starts with a little story (see example below in figure 3.1, all stories can be found in appendix 1 and the questionnaire in appendix 2) that introduces the researcher and asks the participants to read the article carefully priorly to answering the questions. At the end there is a button ‘send’ and when the participants click that a screen with ‘thank you for participating’ shows. The questionnaires are analysed with the statistical software SPSS. In chapter 4 these analyses are discussed in detail.

Image 3.1: Example of story used in the questionnaire

3.5 Testing of the Variables

A questionnaire is used to investigate the response of the consumers to the manipulated variables. All scales that are used are measured on 7-point Likert scales. There are scales that are constructs of several items but also scales that are dimensions of its own.

3.5.1 Brand Attachment

Brand attachment is measured with the attachment scale as presented in Park et al (2010), with just some slight modifications. The construct consists of six items:

‣ I love the brand Land Rover

‣ I like to associate myself with the brand Land Rover

‣ The brand Land Rover fits well with who I am

‣ The image of the Land Rover fits my image

‣ I feel personally connected to Land Rover

‣ I think often about the brand Land Rover

Furthermore, it would be likely that a person will be more attached to something that he/she finds beautiful, because he/she will be more likely to identify himself/herself with a beautiful item. Hence, “I think that Land Rover produces beautiful cars” is added to the brand attachment scale.

3.5.2 Thankfulness

Two items are used to measure the variable Thankfulness:

‣ I am thankful to Land Rover for sponsoring Edward Gal/European Championships Dressage

‣ I am thankful to Land Rover for sponsoring Dressage

3.5.3 Consumer Connection

Consumer Connect is measured on a scale that consists of 6 items. The items are based on items that are used by Lings and Owen (2008) and Thomson (2006). The items used by Lings and Owen (2008) are similar to some of the items used by Thomson (2006). Thomson also uses personal meaning and emotional connection in his construct. In this study being a fan of a sponsored object is also part of consumer connection and hence it is presumed that two of Thomson’s items can be added to the Lings and Owen items in the construct consumer connection.

Lings and Owen (2008)

‣ I feel like I am part of the ‘family’ when I watch Edward Gal/European Championships Dressage

‣ I do not feel like I am part of the ‘family’ when I watch Edward Gal/

European Championships Dressage

‣ Edward Gal/European Championships Dressage means much to me personally

‣ I do not feel emotionally connected to Edward Gal/European

Thomson (2006)

‣ I am a big fan of Edward Gal/European Championships Dressage

‣ It would be a pitty if Edward Gal quits riding/if there would not be European Championchips Dressage anymore

3.5.4 Fit

In order to check whether there is a perceived fit between Land Rover and Edward Gal/European Championships Dressage four items are used:

‣ There is a logical connection between Land Rover and Edward Gal/

European Championships Dressage

‣ The image of Edward Gal/European Championships Dressage and the image of Land Rover are similar

‣ Land Rover and Edward Gal/European Championships Dressage fit well together

‣ It is logical to me that Land Rover sponsors Edward Gal/European Championships Dressage

3.5.5 Background variables

It is important to get an idea of what kind of people participated in the research.

This is also necessary to check if a good comparison can be made between the experimental groups. In order to be able to do this it is necessary to collect some background information. The following background variables are included in this study:

‣ Age

‣ Gender

‣ Education

‣ Ownership of a Land Rover

‣ Connection to equestrian sports

‣ Being active in equestrian sports

‣ Connection to dressage

4. RESULTS & ANALYSIS

In this chapter the results of the research are presented and analysed in order to be able to test the hypotheses and answer the research question. Firstly, an overview of the sample is given. Secondly, the reliability of the scales is tested.

Thirdly, the results are presented descriptively. Finally, an analysis of the results and the testing of the hypotheses can be found.

4.1 Sample Impression

In total there are 125 participants (N = 125), equally spread over the four experimental groups and the control group. In this paragraph an impression of the sample is given, based on demographics and involvement in the focal sport.

A compare means analysis indicates that there is no significant difference in age between the groups (p > 0.05). Furthermore, another compare means analysis shows that there is also no significant difference in gender (p > 0.05). A similar analysis for education also proves no significant difference between the groups (p >

0.05). This indicates that the groups are similar, which is positive in terms of being able to compare the differences of the focal variables in this study between the groups. Since there is no significant difference between the groups, their data are joined to get the descriptive results.

The vast majority of participants are female, namely 94.4%. Furthermore, most participants are relative young and highly educated. Approximately, 64% of the participants are younger than 30 years. A large percentage of the participants has a higher education (HBO), namely 37.6%. Additionally, 22.4% of the participants has a university degree.

Graph 4.1: Age of the participants Graph 4.2: Education of the participants

Approximately 88% of the participants does not own a Land Rover, nor has never owned one before. A compare means analysis shows no significant difference between groups (p > 0.05). All participants knew the brand before the study. It is logical that someone who owns a Land Rover most likely feels more attached to the brand than someone who does not own one. Since, the percentage of participants who own one is low and there is no significant difference between the groups, this will not have a large unwanted influence on the study.

4.2 Scale Reliability

4.2.1 Brand Attachment

In order to see if the seven items that are presumed to form the scale for brand attachment form a reliable scale, firstly a factor analysis is conducted.

Since only one factor is wanted in this research, the command to extract only one factor in SPSS is used. The results of the factor analysis show that the seven items factor well since the KMO score is larger than 0.5, namely 0.891. Furthermore, the Bartlett’s test of sphericity is highly significant (0.000) which indicates that the items are correlated.

In order to decide if just one factor can be formed the eigenvalue should be > 1, the total variance > 60% and the individual variance > 5%. Since this is the case, one factor can be formed for Brand Attachment out of the seven items.

A Cronbach’s alpha test is than executed to check whether the factor is ‘strong’

enough. The outcome of this test should be > 0.6. The score for the factor Brand Attachment is 0.939, hence the factor is ‘strong’ enough.

Table 4.1: Brand Attachment scale reliability

Scale KMO Bartlett’s Cronbach’s

Brand Attachment 0.891 0.000 0.939

4.2.2 Thankfulness

The KMO score for the factor thankfulness is relatively low (0.500). However, the Bartlett’s test of sphericity is highly significant (p 0.000) and the Cronbach’s aplha test is high (0.930). Furthermore, the two items explain over 93% of the variance.

Hence, the two items can fall under just one scale; Thankfulness.

Table 4.2: Thankfulness scale reliability

Scale KMO Bartlett’s Cronbach’s

Brand Attachment 0.500 0.000 0.930

4.2.3 Consumer Connection

There are six items that are proposed to form the scale for consumer connection. A factor analysis is conducted to check if one factor can be extracted from the six items.

When all six items are included in the analysis the KMO is > 0.5 and the Bartlett’s test of sphericity is significant. However, the total variance is not > 60%. Hence it is not possible to extract one factor out of all six items.

The item “It would be a pitty if Edward Gal quits riding/if there would not be European Championchips Dressage anymore” has a low factor score and is therefore excluded from the analysis. The other five items are used in a second factor analysis.

The five items factor well (KMO 0.797) and are correlated (Bartlett’s test of sphericity 0.000). Furthermore, with these five items the eigenvalue is > 1, the total variance is > 60% and the individual variance is > 5%. Hence one factor for consumer connection can be extracted from the five items. This factor is strong enough (Cronbach’s alpha 0.862).

Scale KMO Bartlett’s Cronbach’s

Consumer Connection 0.797 0.000 0.862

Table 4.3: Consumer Connection scale reliability

4.2.4 Fit

The four items that form the scale for fit factor well (KMO 0.753) and are correlated (Bartlett’s test of sphericity 0.000). Furthermore, the eigenvalue is > 1, the total variance is > 60% and the individual variance is > 5%. Hence one factor is extracted for fit. The factor fit proves to be strong enough with a Cronbach’s alpha of 0.883.

Table 4.4: Fit scale reliability

Scale KMO Bartlett’s Cronbach’s

Fit 0.753 0.000 0.883

4.3 Descriptive Results

The participants in the experimental groups are asked to indicate to what extent they feel connected to the focal sport (equestrian) and more specific to the dressage part of equestrian sports.

Furthermore, the participants are asked how active they are in the equestrian sports. All participants feel connected to the equestrian sports to some extent, which is a pre-requisite in this study.

M o r e o v e r , t h e v a s t m a j o r i t y o f participants also feel connected to dressage (76%) and are active in equestrian sports (87%).

Additionally, the participants are asked to what extend they find it important that the equestrian sports are sponsored. All most all participants (94%) find this important. Furthermore, 91% of the participants also find it important that the dressage part of the equestrian sports are sponsored. There is no significant difference between the experimental groups.

Besides whether participants find sponsorship important or not, they are also asked if they are thankful to Land Rover for sponsoring the event/rider. The results show that the participants are on average relatively thankful for the sponsorship.

The mean score is 5.28 out of 7.

The participants in the research are neutral (mean score 4.25) about the fit between Land Rover and the event/rider. Previous research indicated that a strong perceived fit has a positive influence on brand attachment. A regression analysis shows that this also is the case for this

research. The adjusted R2 is relatively low (0.147) however the overall model is significant (p < 0.05) and is therefore accepted. The results show that if the score on fit goes up by 1, the score on brand attachment goes up by 0.394 (standardized beta) at a 0.000 significance level. The majority of people give a score between 3 and 5. Just 7% gives a lower score and 16% a higher score. The effect of the perceived fit is more throughly analysed in the hypothesis testing part of this chapter.

The connection that the participants feel to the event/rider is another important factor that is in the predicted conceptual model. The mean score on consumer

Graph 4.3: Connection to equestrian

Graph 4.5: Perceived fit

On the dependent variable brand attachment the mean score is 3.88 out of 7, a medium score. Since brand attachment is the dependent variable, further analyses need to be conducted and the results are presented in paragraph 4.4.

Variable Mean

Thankfulness Fit

Consumer Connection Brand Attachment

5.28 4.25 3.30 3.88

Table 4.5: Overview means

4.4 Hypotheses Testing

In order to be able to answer the research question, the various hypotheses first need to be tested. Several analyses are conducted to get the wanted results.

4.4.1 Main effect

Before the hypotheses testing can be continued, it has to be determined whether the main effect of sponsorship on brand attachment occurs.

In order to investigate if the manipulated variables (sponsored object and perceived intention) have a direct effect on brand attachment, a regression analysis is conducted. Firstly, the variables are dummy coded.

Sponsored object (three options, two dummy’s):

‣ No sponsorship (0 in both dummy’s)

‣ Individual (dummy)

‣ Event (dummy)

Perceived intention (three options, two dummy’s):

‣ No sponsorship (0 in both dummy’s)

‣ Goodwill (dummy)

‣ Commercial (dummy)

Secondly, products are calculated for the interaction effect between sponsored object and perceived intention:

‣ Individual * Goodwill

‣ Event * Goodwill

‣ Individual * Commercial

‣ Event * Commercial

The variables are then ready for the regression analysis. The results show that the sponsored object does not have a significant effect on brand attachment, the overall regression model is not significant (ANOVA 0.238). The same holds for perceived intention, the ANOVA score is 0.231 for this regression model.

In order to see if the interaction terms have an influence on brand attachment, the two ‘separate’ variables and the interaction term are put together in a regression model. The four regression models for the interaction terms also prove to be insignificant (see table 4.7). None of the regression models are significant. Hence, neither sponsored object nor perceived intention has a direct effect on brand attachment. Additionally, there is no effect of the interaction terms. In paragraph 4.4.2 and 4.4.3 the indirect effect of the two variables is investigated, because as explained previously it is expected that the two variables will have an indirect effect.

Variable Influence on Brand Attachment Sponsored object

Perceived Intention Individual * Goodwill

Event * Goodwill Individual * Commercial

Event * Commercial

Model not significant (ANOVA 0.238) Model not significant (ANOVA 0.231) Model not significant (ANOVA 0.880) Model not significant (ANOVA 0.293) Model not significant (ANOVA 0.278) Model not significant (ANOVA 0.556)

Table 4.7: Effect of perceived intention and sponsored object on Brand Attachment

There seems to be no difference in brand attachment between the experimental groups, because the manipulations do not have a significant effect. This is confirmed by a compare means analysis, which indicates no significant differences in brand attachment between the experimental groups (p > 0.05).

Since there is no significant difference between the commercial and the goodwill groups, hypothesis 4 is rejected. A sponsorship perceived as goodwill is not more beneficial than a sponsorship perceived as commercial.

In the case of perceived intention it might be possible that e.g. the commercial intention was not as perceived as commercial and that this explains the fact that there is no significant difference in brand attachment. Therefore, this is checked.

Participants in the commercial experimental groups significantly (p < 0.05) think

more so that Land Rover sponsors solely to its own benefits than the goodwill

experimental groups. This is an indication that the commercial sponsorship was

perceived indeed as commercial. However, in both groups (goodwill and

commercial) participants on average agree with the fact that the event/rider

Experimental Groups Land Rover Benefits* Event/Rider Benefits Goodwill

Commercial

3.82 5.22

4.56 5.30

* Significant difference between groups Table 4.8: Explanation perceived intention

Since there is no significant effect of the manipulations, the experimental groups are joined into one experimental group, which is compared to the control group all through the rest of the analyses. A compare means analysis between the control group and the experimental group will give the outcome of whether or not the main effect occurs. The effect is weakly significant with a p value of 0.100. The mean score on brand attachment is higher for the experimental group than for the control group, and although weakly significant H1 can be excepted. The weak significance is also excepted as a ‘strong’ enough effect in this research to carry on with the further analyses.

Group Mean

Control Experimental

3.47 3.99

**Weakly significant difference in brand attachment p 0.100) Table 4.9: Main effect

4.4.2 Effect of sponsored object on consumer connection

Firstly, the influence of the sponsored object (individual vs event) on consumer connection is investigated. A compare means analysis shows that there is a significant difference between the individual and event groups, the ANOVA statistic is namely significant (p < 0.05). Hypothesised is that consumers will have a greater connection to an individual than to an event. The results are however not in line, the results clearly show that the mean score on consumer connection is higher in the event groups. Hence, hypothesis 3 has to be rejected.

Experimental Groups Mean Individual

Event

2.95 3.64

Table 4.10: Significant difference in consumer connection (p 0.003)

4.4.3 Effect of consumer connection on thankfulness

The effect of consumer connection on thankfulness is investigated with a regression analysis. The model is significant (ANOVA 0.000) and consumer connection proves to have a significant, positive effect on thankfulness. Hypothesis 5 can therefore be accepted; Consumer Connection has a positive influence on thankfulness.

Variable Thankfulness

Consumer Connection

Standardised Beta 0.413

**Significant influence (p 0.000)

Table 4.11: Effect of consumer connection on thankfulness

4.4.4 Moderating effect of fit

Since the experimental groups do not differ significantly in brand attachment, also in this analysis the data of the groups are joined into one experimental group. This means that there is the control group (sponsorship no) and the experimental group (sponsorship yes). The moderating effect is the interaction between the independent variable and the moderator, which is the product of the two:

Sponsorship (S) * Fit (F)

The regression model is first run with brand attachment (B) as the dependent variable, sponsorship and fit as the independent variables. The interaction term is added in a second model and the change in Rsquared is calculated.

There is unfortunately too much correlation in the model and therefore the interaction term cannot be included in the analysis. The term is automatically excluded by SPSS when the analysis is executed. Hence, the moderating effect of fit can also not be investigated. Hypothesis 7 is thus rejected.

4.4.5 Mediating effect of consumer connection

The mediating effect of consumer connection that is expected in this study needs to be investigated in a couple of steps. To start with, three conditions have to be established:

1. The independent variable (sponsorship) predicts the dependent variable (brand attachment)

2. The independent variable (sponsorship) predicts the mediator (consumer connection)

3. The mediator (consumer connection) predicts the dependent variable (brand attachment)

The mediator becomes zero in case of no sponsorship (control group), because there

cannot be a consumer connection to a sponsored object when there is no