Consumers’ sponsor recall within sport sponsoring The effects of fit, fan identification and team performance

Consumers’ sponsor recall within sport sponsoring

The effects of fit, fan identification and team

2

Consumers’ sponsor recall within sport sponsoring

The effects of fit, fan identification and team

performance

by

Maaike Strating

University of Groningen Faculty of Economics and Business

3

MANAGEMENT SUMMARY

To gain more brand awareness and to sell more products or services, companies worldwide use sponsoring as a marketing tool to promote their products and services. According to Westberg, Stavros and Wilson (2011), the brands of both the sponsor and the sponsored company are central to the success of the sponsor relationship. In this research a focus is given to the different influences of sponsorship effectiveness within a sports club. One of these influences is the fit between a sponsor and the sports club. Many studies argue the relevance of fit, for example Speed and Thompson (2000) state in their research that the response of a consumer will be stronger when this consumer perceives a fit between the sponsor and the club. The sponsorship effectiveness will be measured as consumers’ sponsor recall in this research; this variable was measured by aided and unaided recall. A new angle on the sports sponsoring literature is given as two moderating influences are used in this study. First the influence of perceived team performance is measured as this variable is likely to have a positive impact on the sponsoring effectiveness (Ngan, Prendergast & Tsang, 2011). The second moderating variable of this research is fan identification. According to Donahay and Rosenberger (2007) fan identification leads to efficient image transfers of the sponsor message. Combining these four variables, the following research question was formulated:

How does fit influence consumers’ sponsor recall, and what are the moderating effects of team performance and fan identification on this relationship?

In order to test the proposed hypotheses, an online questionnaire was conducted among supporters of a Dutch football club called FC Groningen. The distribution of the respondents was compared to the distribution of a dataset of FC Groningen which contained all the data about the current season ticket holders, and these two datasets correlated reasonable. When testing the moderating variables, it became clear that multicollinearity between these variables occurred. This might have had an influence on the results of the research. When testing the unaided recall, it can be said that almost 94 percent of the respondents knew the main sponsor of FC Groningen during the 2011/2012 season. Looking at the aided recall question, respondents picked on average 3.27 (out of five) correct sponsors of a given list. In order to test the hypotheses of this research, two regression analysis were performed. The first regression analysis tested the relationship between perceived fit and consumers’ sponsor recall. However, unfortunately, this regression was not significant. To test the hypotheses about the moderating variables, a second regression was performed with the variables perceived team performance, perceived fan identification, both the cross products of the moderating variables and the aided recall in order to measure the aided recall. Again, this regression was not significant. Additional test, like segmenting, were performed as none of the proposed variables had a significant influence on sponsor recall. However, these tests neither had a significant outcome.

4

PREFACE

This research report was written as a completion of my master in Business Administration, specialization Marketing Management at the University of Groningen. In this research a literature review and an analysis about consumers’ sponsor recall within sport sponsoring and some other influencing factors like perceived team performance and fan identification can be found.

Last year, during my premaster, I followed the course Sports Economics where a guest lecture was held about sports marketing. I found this a really interesting topic and I knew that I wanted to write my master thesis within a company in order to experience the business. Via prof. dr. Koning I was able to come in contact with FC Groningen. And as FC Groningen belongs to the sub top of the premier league in the Netherlands, it is a very dynamic company which also has many ‘hardcore’ supporters. Therefore I thought that conducting my research at FC Groningen would be very interesting and useful for me. I would like to use the opportunity to thank my supervisor from FC Groningen, Robbert Klaver for the general support and all the other colleagues of FC Groningen who I have worked with for the pleasant and instructive time. In addition, I would also like to thank my supervisors from the University of Groningen. Dr. K.J. Alsem for providing overall guiding and more specifically on the marketing topic, and prof. dr. R.H. Koning for the guidance particular in the field of sports economics.

5

TABLE OF CONTENT

Management summary

1.3 Structure of the research ... 8

2. LITERATURE REVIEW ... 9

2.1 Fit ... 9

2.1.1 Relevance of fit ... 9

2.1.2 Positive effects of fit ... 9

2.1.3 Effects when a sponsorship scores low on fit ... 10

2.1.4 Dimensions of fit ... 11

2.2 Measuring sponsor recall as the sponsorship effectiveness ... 12

2.2.1 Key determinants of sponsor recall ... 12

2.2.2 Sponsor recall as a sustainable competitive advantage ... 13

2.2.3 Placement of the sponsoring message ... 13

2.3 Potential moderating influences ... 14

2.3.1 Team/fan identification ... 14

2.3.2 Demographic characteristics ... 15

2.3.3 Brand loyalty ... 15

2.3.5 Team performance ... 16

2.4 Team performance and fan identification ... 16

2.5 Conceptual model ... 17

3. METHODOLOGY ... 18

3.1 Method & design ... 18

3.2 The questionnaire and the respondenst ... 18

3.3 FC Groningen ... 19

3.4 Measurement scales ... 19

3.4.1 Perceived fit ... 19

3.4.2 Consumers’ sponsor recall... 19

3.4.3 Fan identification & Team performance ... 20

3.4.4 Extra elements in the questionnaire ... 21

3.5 Analysis ... 21

4. RESULTS ... 23

4.1 Sample representativeness and internal validity ... 23

4.1.1 Sample representativeness ... 23

4.1.2 Internal validity ... 24

4.2 Empirical results ... 25

4.2.1 Outcomes of consumers’ sponsors recall ... 25

4.2.2 Additional testing ... 27

4.2.3 Measuring the variation of coefficients ... 28

4.3 Hypotheses testing ... 29

5 CONCLUSION & RECOMMENDATIONS ... 31

5.1 Discussion ... 31

6

5.3 Limitations and directions for future research ... 34

REFERENCES ... 36

APPENDIX I ... 41

APPENDIX II ... 46

APPENDIX III ... 48

7

1. INTRODUCTION

1.1 Introduction

Worldwide, sponsoring is used by companies to promote their products and services. In 2010 a total of $48,7 billion was spend on sponsoring worldwide. And after North America, Europe is the largest source of sponsorship spending (sponsorship, 2011). But how will a sponsorship relation have the best results for a company? Sponsorship has been defined by Meenaghan (1983) as a ‘’provision of assistance either financial or in kind to an activity by a commercial organization for the purpose of achieving commercial objectives. Furthermore, it can be said that sponsoring is a component of public relations. According to Cornwell and Maignan (1998, p. 11) a sponsorship should consist of two critical activities relating to exchange and associations if it wants to be an effective investment: (1) and exchange between a sponsor and a sponsee whereby the latter receives a fee and the former obtains the right to associate itself with the activity sponsored, and (2) the marketing of the association by the sponsor. A sport sponsorship concerns a strategic business-to-business relationship between a sport entity and a sponsor for a mutual benefit (Farelly & Quester, 2005). According to Westberg, Stavros and Wilson (2011) the brands of both the sponsor and the sport entity are central to the success of a sponsor relationship. Sponsorships can be directed at consumers, the community, the government, employees, channel members and financial institutions (Gardner & Schuman, 1988) and in addition they can be directed at much different kind of events, like for example sport events.

According to Amis, Pant and Slack (1997), a sponsorship agreement should be considered as a strategic investment. Moreover, a sponsor should handle their sponsorships as a resource which can be developed to a sustainable competitive advantage, as a sport sponsorship might effectively enhance a company’s reputation and image (Amis, Slack & Berrett, 1999). Possible results of sponsoring are that it can have an influence on consumer awareness, consumers’ purchase intentions and attitude towards the sponsor (Rifon et al., 2004). A key objective from the sponsor, are the positive effects on their corporate brand image by sponsoring a sport organization or event (Javalgi, Traylor, Gross & Lampman, 1994). Therefore the perceptions of the sponsor’s brand in the mind of consumers and other stakeholders might be boosted by the use of the (brand) image of the event or sport organization (Gwinner & Eaton, 1999; Farrelly, Quester & Burton, 2006).

However, sponsoring can also have less positive effects for the sponsoring firm. A study of Yoon and Gurhan-Canli (2003) reveals that when consumers receive information which is inconsistent with their prior knowledge, this may cause consumers to question the underlying motives for the sponsor relationship. And after a while this may even lead to a weakened brand image of both partners. Furthermore, when a team, event or organizations attracts negative publicity, this might have a negative impact on the sponsorship relation (Amis, Slack & Berrett, 1999). Hence, when firms decide to sponsor a sports club, there is always a certain level of risk which is associated with, inter alia how the club will perform in the future. Additionally, a sponsoring firm can never be exactly sure how the club is perceived by those to whom the sponsor is trying to appeal (Amis, Slack & Berrett, 1999). However, a fit between a sponsor and an organization can have positive effects with respect to the sponsorship effectiveness (Johar & Pham, 1999). Therefore, a lot of existing literature about (sport) sponsoring refers to the importance of fit between the sponsor and the sponsored organization.

1.2 Research question

8 Fit refers to ‘’the extent to which a sponsor and an event are perceived as being similar, whether that similarity is derived from functionality or image’’. Sponsor recall can be defined as ‘’a measure of memory, and it can be measured by aided and unaided recall’’ (Lucas, 1960). The second objective is to determine what the effects of sportive results and fan identification are on the relationship between perceived fit and sponsor recall. The problem statement of this research is therefore:

How does fit influence consumers’ sponsor recall, and what are the moderating effects of perceived team performance and fan identification on this relationship?

This research is partly based on existing studies. However, by using two moderating effects, a new angle on the sport sponsorship literature is given.

1.3 Structure of the research

9

2. LITERATURE REVIEW

Based on existing literature, it should be determined what sort of influences a perceived fit in sponsoring might have on the sponsor recall of consumers. Therefore a literature review of the different types of fit relating to sponsor recall will be given. Then, a literature review of the impact of (perceived) team performance and fan identification on the relationship between fit and sponsor recall will be given. This section will end with the hypotheses and conceptual model for this research.

2.1 Fit

According to Olson and Thjømøe (2009) the sponsoring effectiveness might be influenced by the type of communication goal, the level of familiarity of the consumer, the type of exposure the sponsor receives, and the fit of the sponsor with his sponsored object. In this section, the variable fit will be discussed. First the relevance of fit in sponsoring will be shown, and after that the positive effects of fit will be revealed, whereas in the last part of the section is described how to handle a low fit sponsor relationship.

2.1.1 Relevance of fit

A frequently studied concept in sponsorship research is the notion of fit (also referred to as match, relatedness or congruence) between a club or firm being sponsored and the sponsor. Fit refers to the extent to which a sponsor and an event are perceived as being similar, whether that similarity is derived from functionality or image (Coppetti et al., 2009). A functional fit can be explained as whether the two parties fit from a functional perspective, for example whether the brand/ product of the sponsor can be used by the participants of the event. An image fit relates to the associations and image the sponsor and the sponsored party share (Gwinner & Eaton, 1999). The concept of fit is not only true for sponsorship relations. Völckner and Sattler (2006) argue that in the case of brand extension success, a fit between the parent brand and a product extension might be the most important driver of a successful extension. Furthermore, in the case of celebrity product endorsement, several authors state that there should be a high fit between the endorsed product and the celebrity endorser (Atkin & Block, 1983; Lee & Thorson, 2008; Koernig & Boyd, 2009; Liu et al., 2010). Furthermore, the influence of fit is moderated by some variables, like consumers’ level of brand consciousness (Nan & Heo, 2007), the affinity consumers hold for the social causes and consumers’ perceptions of the firm’s motives (Barone, Norman & Miyazaki, 2007).

2.1.2 Positive effects of fit

10 In addition, a study of Becker-Olsen and Hill (2006) shows that when a sponsorship is perceived as having a high fit, this might lead to building brand relationships, which ultimately lead to more brand engagement. Furthermore, a high-fit will positively influence volunteerism, a higher willingness to recommend, and individual financial contributions (Becker-Olsen & Hill, 2006). Other research shows that there occur differences in the relative effectiveness of a sponsorship, depending on the communication goal of the sponsor. For example, a study of Olson & Thjømøe (2009) showed that high fitting brands might have a higher conversion rate with respect to brand liking compared to brand recognition.Finally, when a product is congruent with the sponsored sport, this is a positive indicator of perceived sponsorship success, as reasonable related sponsorships are positively related to stock price changes (Cornwell, Pruitt, & Clark, 2005). Speed and Thompson (2000) state that due to a high level of fit between a sponsored event and a sponsor, there will be a stronger positive association between personal liking for the event, and the sponsorship response.

The results of a study of Olson and Thjømøe (2011), suggests that when a firm wants to benefit from a high fit, they should sponsor objects which is equally liked by the target market, which is relevant to the participants and has an audience that overlaps the target market. Furthermore, audience similarity, use by the participants, geographic similarity and attitude similarity, are all significant predictors of overall fit, whereby the use of participants and audience similarity are the most important predictors (Olson & Thjømøe, 2011). Furthermore, a consumer response to a sponsorship is stronger when the consumers perceive a fit between the sponsor and the event (Speed & Thompson, 2000, p. 235). Furthermore promotions who communicate the fit between the sponsor and the event will lead to an increase in sponsorship response (Speed & Thompson, 2000). Therefore, it is important for firms of events to understand the attitudes which are held about the event or firm, as this is a critical element in selecting sponsors and in adding value for these sponsors (Speed & Thompson, 2000).

2.1.3 Effects when a sponsorship scores low on fit

The research of Coppetti et al. (2009, p. 17) points out that many companies may not have logical, native links to sports, events and causes and may therefore end up sponsoring properties that are not a natural match. So, besides the fact that a high fit will have positive effects, there are several studies which show how to handle a perceived low-fit sponsoring (Coppetti et al., 2009). According to Coppetti et al. (2009), the less positive results of a low-fitting relationship can be counteracted by the use of articulation and through audience participation. Articulation (explicitly articulating the basis and the meaning of the sponsorship relationship) of the sponsorship relationship may lead to more favorable brand attitudes, an improved evaluation of the sponsorship and an increased image transfer (Coppetti et al., 2009). These results are in line with the study of Olson and Thjømøe (2011) who suggests effectively communicating the sponsoring to overcome the poor fit, as this might increase the overall fit perceptions of consumers by more than 30%. Furthermore, by using audience participation, the evaluations of incongruent sponsorship relations can be improved. On-site visitors are likely to have a more positive evaluation of the sponsorship due to the participation in attractive sponsor activities (Coppetti et al., 2009). Therefore, the study of Coppetti et al. (2009, p.25) suggests companies to create memorable sponsorship experiences if those firms want to increase the responses to sponsorships that are incongruent.

11 These results are in line with the outcomes of a study of Simons and Becker-Olsen (2006, p.154) who found that the unexpectedness of a low fit leads to increased elaboration. In addition, the study of Simons and Becker-Olsen (2006) also shows that sponsorship recall, attitude towards the sponsorship and firm equity are all likely to be higher when there is a created fit instead of a low fit between a firm and the sponsor. Created fit is here explained as product-related donations which evoke shared associations and with messages that explain how the firm fits the cause (Simons & Becker-Olsen, 2006, p. 161).

2.1.4 Dimensions of fit

Several studies show that there are different dimensions of the variable fit (see table 1, cited by Pentecost & Spence, 2009). Most of these researchers argue that that there are two different type of dimensions; image and functionality fit (Gwinner & Eaton, 1999; Grohs, Wagner & Vsetecka, 2004). In addition, there are also some authors who make a distinction in attribute fit, by dividing this dimension of fit in benefit and usage attribute fit (Dean, 1999; Keller, 2003).

Table 1: Summary of empirical findings of dimensions of fit in sponsoring.

Author(s) Definition of fit. Dimensions of Fit

McDaniel (1999)

Match-up:

Using advertising based schema fit the author states: “One of the implicit principles in the match-up hypothesis is that consumers have memory based expectations of the attributes embodied by celebrities, brands, and product categories”. (p 168)

Image and Functional:

“Advertising response is influenced by a perceived match (or similarity) between an endorsers image attributes, and/or the function of a product (as moderated by product category involvement)” (p 167).

Gwinner and Eaton. (1999)

Congruence/ similarity:

“It can be argued that congruent event-brand information in the form of either functional or image based similarity will lead to enhanced image transfer.” (p 49)

Functional or image:

“Functional based similarity can occur when the sponsored brand is actually used by participants during the event. (p. 49). “Image based similarity has been described as occurring when the image of the event is related to the image of the brand” (p 49)

Dean. (1999)

Linkage:

Using Heider’s (1958) Balance Theory a “belief is out of balance and unstable if a lowly valued object is linked with a highly valued object.” (p 4).

Benefit/usage:

Quality “is defined as an overall judgment of a brands excellence or superiority of performance (with respect to its intended purpose) relative to alternative brands” (p2).

Image:Esteem is “the degree to which the brand is held in high regard, is trusted and is respected by its valued customer”(p 2).

Speed and Thompson (2000)

Congruence:

“The response to a sports sponsorship is proposed to be affected by (1) attitudes towards the event, (2) attitudes towards the sponsor, and (3) perception of congruence between sponsor and event.” (pp 227-228).

Image:

12

Keller (2003)

Associations

“Events have their own set of associations that may become linked to a sponsoring brand under certain conditions.” (p 381)

Image and Benefit/usage:

“Events can be chosen on attendee’s attitudes and usage regarding certain products or brands” (p 317)

Ruth and Simonin. (2003)

Congruency:

“Sponsorship research has investigated congruence effects in the past, where the focus is on understanding the ‘fit’ between the sponsor and the event itself.” (p 22)

Functional and/or Image:

“The transfer of image from the event to the brand was higher when the event and the sponsor were congruent in either functionality or image” (p 22) Grohs, Wagner and Vsetecka. (2004) Link/synergy/similarity:

“Scientific literature has used numerous words to describe the fit between a sponsor and a sponsored activity, such as synergy, similarity, or link.” (p 122)

Image and Functionality:

“Generally, most authors distinguish between a functional fit and an image fit” (p 122)

Rifon, Choi, Trimble and Li (2004)

Congruence:

“The study presented in this paper develops and tests a theoretical explanation for the effects of congruence on consumer attitude towards the sponsor of a cause.” (p 29)

Functional and Image:

“…direct relevance as ‘functional based similarity’ which occurs when the sponsor’s product is used during the sponsorship event, and indirect relevance as ‘image based similarity’…” (p 30).

2.2 Measuring sponsor recall as the sponsorship effectiveness

A sponsorship relation can have several results for the sponsoring firm, like an increase in consumer purchase intention, sponsor recall, or a change in attitude to the sponsored brand (Johar & Pham, 1999; Simons & Becker-Olsen, 2006). In this research, consumers’ sponsor recall will be used to test the sponsoring effectiveness. Sponsor recall can be defined as ‘’a measure of memory’’, and it can be measured by aided and unaided recall (Lucas, 1960). The effectiveness of a sponsorship is generally examined in either terms of consumer psychology or financial outcomes (Kim, 2010). These financial outcomes are mostly measured in terms of stock market return, while consumer psychology is commonly measured by consumers’ purchase intentions and sponsors recall (Kim, 2010). In this research the consumer psychology outcomes will be measured.

2.2.1 Key determinants of sponsor recall

13 Exposure effects which may be conscious and non-conscious, has been shown to positively influence the awareness and liking of the exposed product or brand (Janiszewski, 1993). This is an important finding, as many consumers do not focus their attention to the commercial messages but to the game or event they are visiting. Since these corporate exposure effects occur, the advertised brands/ products will become familiar and this familiarity might positively increase consumers’ sponsor recall (Wakefield, Becker-Olsen & Cornwell, 2007, p. 63). The last predicting variable of sponsor recall strongly correlates with the involvement of the consumer. When a consumer is highly involved, they will probably attend more games or events, which ensure that this consumer is more exposed to the commercial messages. Furthermore, fans are believed to be more emotionally involved, and according to Cross (1999), information that is seen as emotionally affecting is better remembered. However, this is only true for prominent and related sponsors (Wakefield, Becker-Olsen & Cornwell, 2007).

A study of Mason and Cochetel (2006) shows that when there is a change in sponsor, the original sponsor maintains high awareness levels among the audience particularly with the previously sponsored event. In other words, a long-term sponsorship may support long-term brand awareness among the audience. Mason and Cochetel (2006) also argue that when there is a change in sponsor, the new sponsor cannot expect the audience to assume an automatic assumption of ‘’ownership’’ by sponsoring an event or firm. Therefore, Mason and Cochetel (2006) recommend the new sponsor to frequently promote the similarities between the sponsor and the firm values, in order to make the meaning of the sponsorship clear to the consumers. Furthermore, more and more sporting events become ‘’greener’’ and have a focus on social responsibility (HBR blog, 2010). This can be an interesting item for firms, as a study of Barone et al. (2000) shows that that some consumers might be willing to pay a higher price, to support a company which is being perceived as being socially responsible.

According to Bennett (1999) a sponsorship is not only effective in enhancing brand and/or company awareness and recall, but also in creating the perception among supporters of the widespread use (and therefore, desirability) of the products of sponsoring firms. Furthermore, false consensus (i.e. the belief that team sponsors' brands are purchased by far higher proportions of, first, fellow supporters and, second, members of the general public than is actually the case) is an important outcome to team sport sponsorship (Bennett, 1999).

2.2.2 Sponsor recall as a sustainable competitive advantage

A study of Amis, Slack and Berrett (1999) shows that a firm can create a sustainable advantage when they integrate their sponsoring activities with their other marketing initiatives in order to convey a consistent and clear message to their consumers. When a firm wants to create a sustainable competitive advantage, a sponsorship needs to improve the perceived image of the sponsoring firm (Amis, Slack & Berrett, 1999). Moreover, when an entity and a sponsor become interrelated in the eyes of the consumers, the sponsor will benefit from the image and attitudes held towards the sponsored business (Amis, Slack & Berrett, 1999).

2.2.3 Placement of the sponsoring message

14 Next to this, there are differences in sponsor recall when comparing field sponsorship and television sponsorship. Field sponsorship is represented by ‘’messages that are limited to the brand name or to a few words summarizing the brand’s positioning platform’’ (Pham & Vanhuele, 1997, p. 407). And the definition of television sponsorship is ‘’a system whereby an advertiser associates its name or trade name with a program or its promotion, excluding any direct or indirect commercial promotion of its products or services (Lardinoit & Derbaix, 2001, p. 168). The study of Lardinoit and Derbaix (2001) shows that television sponsorship outperforms field sponsorship in the area of memorization, as television sponsorship deals with recency and primacy phenomena which occur since the sponsors’ messages are shown directly before and after the sport event. However, the intensity of the affective reactions which are simulated, the nature of the product, the historical links between the brand and the sponsorship and the salience of the brand are other factors which drive the bulk of the memorization (Lardinoit & Derbaix, 2001).

In sum, there are several factors which can influence consumers’ sponsor recall; perceived fit of the sponsorship is one of them. Previous research suggest further investigating this topic as sports sponsoring is used more often nowadays and since there are conflicting outcomes in current literature (Irwin & Asimakopoulos, 1992; Simons & Becker-Olsen, 2006). As a result, this discussion gives rise to the first hypothesis:

H₁: Perceived fit between a firm (sponsor) and a sports club will positively affect consumers’ sponsor

recall.

2.3 Potential moderating influences

Several factors might have a moderating influence on sponsoring effectiveness. These moderating influences are generally individual and group factors like: past experience with the sponsor/familiarity, knowledge of the sponsor, consumer involvement, arousal and social alliance (Cornwell, Weeks & Roy, 2005). In this section, various moderating influences will be shortly described.

2.3.1 Team/fan identification

15 This might suggest that higher involvement leads to more exposure to the sponsorship, which might lead to a better sponsor recall (Shank & Beasley, 1998). In addition, Ko et al. (2008) found that involvement may positively influence corporate image, sponsorship awareness and purchase intentions. And according to a study of Wakefield, Becker-Olsen and Cornwell (2007), consumers who are highly involved are likely to pay attention to sponsorship exposure, as they are more willing to engage in active information processing.

2.3.2 Demographic characteristics

Several studies showed that demographic characteristics of consumers might have an influence on sponsorship effectiveness (Kinney, McDaniel & DeGaris, 2008). The variables gender and age might have a moderating influence on sponsoring relationships (McDaniel, 1999; Kinney, McDaniel & DeGaris, 2008). Also in sport sponsoring literature, an influence of the variable age can be found.

A study of McDaniel (1999) found that females might have more favorable responses towards sponsoring communication than males. This corresponds with a study of Nufer and Bühler (2010) which argues that gender might have an influence on sponsor recall. However, in contrast with the study of McDaniel (1999), the research of Nufer and Bühler (2010) show that male supporters are more likely to correctly recall sponsors. Furthermore, a study of Kinney, McDaniel and DeGaris (2008) found that males typically have a higher ability with respect to sponsor recall. Moreover, their study also found that the older the consumers are, the less likely these consumers are to recall the sponsor (Kinney, McDaniel & DeGaris, 2008).

2.3.3 Brand loyalty

A study of Levin, Beasley and Gamble (2004) studied the effect of sponsorship of supporters’ loyalty towards the sponsoring brands. However, there are no studies which examined the moderating effect of brand loyalty on sponsorship effectiveness. Oliver (1999, p. 34) defines brand loyalty as ‘’a deeply held communication to rebuy or repatronize a preferred product/service consistently in the future, thereby causing repetitive same-brand or same brand set purchasing, despite situational and marketing efforts having the potential to cause switching behavior’’. Then, brand loyalty can be divided into two types of loyalty, namely attitudinal and behavioral loyalty. Attitudinal loyalty includes a degree of dispositional commitment in terms of some unique value associated with the brand, while behavioral loyalty refers to repeated purchases of the brand (Chaudhuri & Holbrook, 2001). According to the results of Levin, Beasley and Gamble (2004) fans of a sports club may exhibit stronger brand loyalty to sponsoring brands than non fans. This is mostly true for the attitudinal loyalty component of brand loyalty. Furthermore, the loyalty towards the sponsoring brand might be higher for consumers who score high on fan identification (Levin, Beasley & Gamble, 2004). Further research shows that consumers who socially interact with other consumers often demonstrate a strong loyalty towards the sponsoring brands (Rosenbaum, Ostrom & Kuntze, 2005).

16 2.3.4 Familiarity

Another variable which might influence sponsoring effectiveness is the level of brand familiarity of consumers. Brand familiarity relates to ‘’the extent of a consumer’s direct and indirect experience with a brand’’ (Campbell & Keller, 2003). This familiarity is built upon current brand associations. According to Campbell and Keller (2003), processing of commercial messages is different when a brand is familiar versus when it is unfamiliar to the consumer. Moreover, there are a number of studies who argue that consumers who have favorable experiences with the sponsoring firm are more likely to develop favorable attitudes towards this sponsoring (Speed & Thompson, 2000; Olson, 2010; Walraven, Bijmolt & Koning, 2011). The results of Walraven, Bijmolt and Koning (2011) show that familiarity with the sponsor might positively relate to consumers’ sponsor recall. Furthermore, a study of Pope and Voges (2000) shows that familiarity with a brand might relate to higher purchase intentions.

2.3.5 Team performance

Team performance, which refers to ‘’one’s level of attachment to, or concern about, a particular sports team’’ (Wann & Branscombe, 1993), might have an influence on the sponsoring effectiveness. Pope, Voges and Brown (2009) state that sports team sponsoring should be merely focused on successful teams. According to Ngan, Prendergast and Tsang (2011), team performance might have a positive effect on consumers’ purchase intentions to the sponsor’s product; this is especially true for winning teams. Furthermore, the purchase intentions were stronger for teams which contained a ‘’star’’ and in the case of fervent fans. A study of Pope, Voges and Brown (2009) found that positive team performance may have a positive effect on consumers’ perceptions of sponsoring brand quality. Team performance in its entirety can create strong associative linkages between team performance and brand quality (Pope, Voges & Brown, 2009).

2.4

Team performance and fan identification

Following the moderating variables mentioned in section 2.3, the focus of this research will be on the moderating variables team identification and team performance. The assumption is that both variables will have a moderating (positive) influence on the relationship between perceived fit and consumers’ sponsor recall.

As the study of Ngan, Prendergast and Tsang (2011) show, team performance might have a positive impact on sponsoring effectiveness. Not many studies have addresses the topic of team performance/ sportive results, and how these results can have an impact on sponsorship effectiveness. However, conflicting results have been found. According to a study of Walraven, Bijmolt and Koning (2011) the successes of a club appears to have a slightly negative effect on sponsor recall. This might be explained by a higher level of excitement during well played games, which lead to distraction of the commercial messages and a lower recall. However, this does not have to entail that sportive successes hinder favorable consumer responses, as consumers also unconsciously collect sponsorship messages (Janiszewski, 1993).

17 As the second moderating variable, fan identification is chosen as several studies showed that higher fan identification leads to better effectiveness of sponsorship (Bennett, 1999; Donahay & Rosenberger, 2000; Kinney, McDaniel and DeGaris, 2008). Furthermore, as Wakefield, Becker-Olsen and Cornwell (2007) state that fan and/or team identification is one of the four key determinants which are influencing sponsor recall, this variable might be very relevant for the results of this research. And as Shank and Beasley (1998) show, higher involvement or higher fan identification might lead to more exposure to the sponsorship, which might in the end lead to a better sponsor recall.

If the outcome of a match influences the sponsor recall of consumers’, as Walraven, Bijmolt and Koning (2011) shows, it is important for sponsors and potential sponsors to pay attention to these sportive outcomes. However, this area is one which can not be influenced by neither the sponsoring firms nor the management of the sports club itself. The assumption in this research is that the better the team performance is, the higher the sponsor recall of consumers will be, due to the unconsciously collection of sponsoring messages. Furthermore, as Wakefield, Becker-Olsen and Cornwell (2007) states, involvement may positively influence active information processing. Therefore, the second moderating variable is assumed to positively influence consumers’ sponsor recall when these consumers score high on fan identification. These themes give rise to the second hypotheses of this study, which is designed to offer the opportunity to test the relationships inherent to the conceptual model:

H₂A: Well perceived team performance will increase consumers’ sponsor recall when there is a

perceived fit between the sponsor and the sports club.

H₂B: High perceived fan identification will increase consumers’ sponsor recall when there is a perceived

fit between the sponsor and the sports club.

2.5 Conceptual model

This research will investigate what the effect of fit versus non-fit will be on consumers’ sponsor recall and what the moderating effects of sportive results of a club and fan identification will have. The hypotheses which are mentioned in the literature review come together in the conceptual model (figure 1).

Figure 1: the conceptual framework

H1

H2B +

+

Perceived fit Consumers’ sponsor recall

Perceived fan identification Perceived team performance

18

3. METHODOLOGY

In this section, the method of collecting data is given, as well as the different procedures for the various variables. Furthermore, the different elements of the questionnaire are discussed. A copy of the questionnaire used in this research can be found in appendix I.

3.1 Method & design

In this research the hypotheses, which were derived from the conceptual model, will be tested by using an online questionnaire. This research method is chosen as the variable ‘’consumers’ sponsor recall’’ can only be measured in a real-life setting. A questionnaire is therefore the most appropriate research design.

3.2 The questionnaire and the respondents

As this research will examine how the conceptual model will work for a football club as FC Groningen, supporters of FC Groningen were questioned. A brief introduction about FC Groningen can be found in section 3.3. The supporters were approached via the social media of FC Groningen. The goal with respect to the number of respondents was to receive at least 100 completed surveys. When supporters are on the Facebook or Twitter page of FC Groningen they saw a short introduction of the research plus a link, which opened the survey. As we had to deal with supporters of a Dutch football club the questionnaire was held in Dutch to make the survey more accessible. With respect to the procedure, the questionnaire was put on the social media of FC Groningen on Friday the 20th of April. , and the supporters of the club had two weeks to complete the questionnaire.

This questionnaire was divided into six parts. The first part involved questions about the sponsor recall of the respondents, which are divided into aided and unaided recall. The second part was about measuring respondents’ general impression of Essent and their product preference towards Essent. Then, the third part was set up on behalf of Essent as they, as the main sponsor, wanted to know how supporters think of the sponsorship between Essent and FC Groningen. The fourth part of this survey measured respondents’ perceived fit between the sponsor and FC Groningen. The fifth part involved the moderating variables, with questions about the level of fan identification and about the perceived team performance. And the final part of the questionnaire involved some general demographic questions. The questionnaire had the same format for every respondent. In table 2, a summary of the questionnaire can be found, which contains all the different items and some sample questions.

Table 2: Elements and example questions of the questionnaire Part Measuring elements: Number of

questions

(Example) question

1 Unaided recall 1 Which company is the main sponsor of FC Groningen

during the season 2011/2012?

Aided recall 1 Which five companies of the given list were in your

opinion sponsors of FC Groningen during the season 2011-2012?

2 General impression of Essent

4 Rate your overall impression of the firm Essent that sponsors FC Groningen. Unfavourable - favourable Product preference of

Essents’ products

19

3 General message of the sponsor

relationship between Essent and FCG

1 What do you think the general message of the cooperation between Essent and FCG?

Keywords of the cooperation between Essent and FCG

1 Favourable, relevant, green

4 Perceived fit 5 The image of the event and the image of the sponsor are

similar.

5 Perceived fan identification

4 I consider myself as a loyal fan of FC Groningen. Perceived team

performance

6 How important is it for you that FC Groningen scores high up in the Premier league?

6 Demographics 5 Age, gender, level of education, being a season holder

ticket of FCG and playing football him or herself

3.3 FC Groningen

The sports club which was used as a stimulus for this research is FC Groningen. FC Groningen is a football club which plays in the Dutch premier league and is located in the Northern part of the Netherlands. The club was founded on June 16th 1971. The club is known as one of the Dutch clubs who has the most avid supporters. The average duration of a supporter who is a season ticket holder is 13.8 years, this can be considered as high when compared to other clubs (the average is around ten years). A well-known nickname of the club is ‘’de Trots van het Noorden’’ which can be translated as ‘’the pride of the North’’ (FC Groningen, 2012). Furthermore, the current results (season 2011/2012) and the position on the ranking are worse compared to the previous seasons.

Since the beginning of the season 2011/2012 is Essent the main sponsor of the club, this relationship is signed for the upcoming three seasons. Along with FC Groningen their aim is to emit a green partnership (Essent, 2011). So, the supporters of the club, the current team performance and the main sponsor of FC Groningen made this club a very interesting and attractive club to use for this research as all the effect which this research wants to measure are present.

3.4 Measurement scales

In the following sections, the measurement scales for the variables which are used in the research can be found.

3.4.1 Perceived fit

The independent variable in this research is perceived fit. The perceived fit between FC Groningen and the sponsor was measured by asking five statements with a 7-point Likert scale. These statements have also been used in a research of Speed and Thompson (2000), and they refer to abstract notions of fit to avoid respondents implying a particular definition of fit for themselves. This variable is expected to positively influence the sponsor recall of consumers.

3.4.2 Consumers’ sponsor recall

20 First the unaided recall was checked by asking the respondents who the main sponsor of FC Groningen during the 2011/2012 season is. Then, comparable to the study of Gwinner and Swanson (2003), respondents were asked to identify sponsors of FC Groningen from a given list, in order to check the aided recall. Ten firms are given from different types of industries (e.g. energy company, marketing). Half of these organizations are actually sponsors of FC Groningen, while the other half are companies which are in fact (direct) competitors of the real sponsors. Respondents than had to select five firms out of this list, of who they think, are a sponsor of FC Groningen. These answers were measured by checking whether the respondents correctly identified the sponsors and non-sponsors from the list. In order to see how the respondents scored on the aided recall question, the following equation was prepared:

With this equation, the descriptive statistics about the aided recall can be measured in order to find out how the respondents performed. In this equation M stands for the average. The 2.1 in this equation stands for the guess probability which the respondents have when they randomly select five of the companies which were in the list. Normally this guess probability would be 2.5 when the respondents would select five out of the ten firms. However, as not all respondents selected five companies, the guess probability has been calculated to this specific dataset. The calculations of this guess probability can be found in table 3, appendix II. The 0.0696 in the formula stands for the standard deviation of the number of correct answers, the calculation of this standard deviation can be found in table 4, appendix II.

Furthermore, the Signal Detection Analysis was used to measure respondents’ sensitive scores (A’). This analysis was invented by Green and Swets (1966) and can be used to control the guessing probabilities of respondents. With the sensitivity scores, it can be said how accurately the respondents identify the sponsors of the non-sponsors. This score is calculated by the following formula:

Here ‘’X’’ stands for the probability of false alarms, while ‘’Y” stands for the probability of a hit. As stated before, respondents needed to select five out of ten companies from a list, of which five of them are sponsors and the other five are not. Therefore, four different types of responses were possible, see table 5: (1) a hit, the correct identification of a sponsor, (2) a miss, an incorrect identification of a sponsor to be not a sponsor, (3) a false alarm, the incorrect identification of a non-sponsor as a sponsor, and (4) a correct rejection, a the correct identification of a non-sponsor as a non-sponsor.

Table 5: the different types of responses of the Signal Detection Analysis ‘’Incorrect’’ Respond ‘’Correct’’ Respond

Sponsor Miss Hit

Non-sponsor False Alarm Correct Rejection

3.4.3 Fan identification & Team performance

21 This scale has been widely used, and mostly in the sport marketing literature (Wann, 1995; Wann, 1996; Bernache-Assollant, Bouchet & Lacassagne, 2007). This scale consists of six items, which are all rated on a 7-point Likert scale; with lower scores indicate a lower level of identification, while higher scores refer to higher levels of fan identification. Furthermore, in order to use this scale for this specific research, the type of sport was changed to football in these questions.

There are a number of ways to measure a team’s performance. However the goal in this research is to measure the perceived team performance of the supporters. As a scale which measures perceived team performance could not be found, questions had to be prepared. Ultimately six questions were set up to measure the consumers’ perceived team performance variable within the current season. These items are all rated on a 7-point Likert scale.

For these two moderating variables, also a cross product was set up in order to test for the effect of the moderating variables on the relationship between fit and sponsor recall. These cross products were also used in the second regression analysis.

3.4.4 Extra elements in the questionnaire

As has been stated in the method section of this chapter, there were some extra questions about the relationship between Essent (which is the main sponsor of FC Groningen) and FC Groningen. These questions were set up on behalf of Essent as they would like to get more information about how supporters think of this relationship. First a general, open question was asked, in order to see what the respondent think is the main message of the cooperation between both parties. Then, a second question, which contained 25 keywords, was set up. The respondents were then asked to pick one or more of these keywords to characterize/express their feelings with respect to the relationship between Essent and FC Groningen.

Furthermore, the attitude towards the sponsor was measured by asking the respondents to rate their overall impressions towards the sponsor. In order to measure the attitudes of the respondents, a seven-point semantic differential scale of Gwinner and Swanson (2003) was used. An additional question about the product preference of Essent was asked. Finally, some general, demographic questions were asked and whether the respondents were a season ticket holder of FC Groningen and whether they play football themselves.

3.5 Analysis

23

4. RESULTS

Before the questionnaire was put online, the survey was pretested. After a total of seven respondents filled in the pretest, there were no more comments and changes to be made. These seven respondents consisted of three fellow students and four managers from FC Groningen. Before showing the results of this research, it needs to be said that 232 respondents have completed the online questionnaire. This number of respondents is far beyond the conditional 100 respondents. However, 17 respondents were deleted from the results as their results were incomplete, so the results are based on a total of 215 respondents (92.7 percent of the total sample). However, 58 respondents did not complete the questions about the independent and moderating variables. Therefore, a total of 157 respondents were used for the actual analysis in order to test the hypotheses.

In the first section of this chapter, the distribution of the respondents will be shown. Then the empirical results will be shown. The empirical results section will show the outcomes of the aided recall question, some additional testing and the measurement of the sensitivity scores. This chapter will end with testing the hypotheses, in order to show if respondents actually have an increased sponsor recall when they perceive a fit between FC Groningen and the main sponsor Essent. Subsequently the influences of the moderators will be explained.

4.1 Sample representativeness and internal validity

4.1.1 Sample representativeness

In order to see how the distribution of the respondents was divided, a cross tabulation of each passive variable (age, gender and level of education) was made. To examine if this research represents real data and to check whether the distribution of the data corresponds with the real population, the demographic outcomes of the chosen population are compared to a dataset of FC Groningen. This dataset consist all information about the season ticket holders of FC Groningen during the season 2011/2012, were all 15751 season ticket holders are present.

In table 6, the outcomes of the sample population and the real population (the dataset of FC Groningen) can be found. When the demographic variables are compared to the real data, it can be said that the age of the respondents is reasonable similar to those of the real data. In order to analyze whether the average age of the respondents is significantly different to the average age of the real dataset, a one-sample t-test was performed. This one-one-sample t-test was significant t (152) = -5,394, p = ,000. So it can be said that the average age of the respondents is significantly differs from the average age of the real data set.

The gender of the respondents was almost exactly equal to those of the real data. But just as with the variable age, a one-sample t-test was performed to test whether these outcomes are significantly different. This one-sample t-test was non-significant t (154) = -,281, p = ,779. So it can be said that the gender distribution of the respondents is significantly equal to the gender distribution of the real data set.

24 Table 6: representativeness of the sample

Items Sample population Real dataset of FC Groningen

Age 34.79 year 41 year

Gender 82.6% male, 17.4% female 83.2% male, 16.8% female

Level of education 46.8% HBO-level Unknown

Season ticket holder 56% is a season ticket holder1 100%

Playing football 34.2% played football Unknown

4.1.2 Internal validity

In order to analyse the internal validity and the correlations between the items to minimize the amount of analyses, the Cronbach’s Alpha and the Pearson correlation test were performed. In table 7, the outcomes of the p-values of the Pearson correlation tests were given. The actual correlation coefficients of the Pearson correlation tests can be found in table 8 until 10 in appendix III. Looking at table 7, it can be said that all the items about fit, team performance and fan identification had a Cronbach’s alpha far above the additional 0,7. This implies that these questions can be aggregated, and therefore three new variables, namely perceived fit (IV), perceived fan identification (MV) and perceived team performance (MV) were created.

Table 7: Internal validity and correlation

Items Pearson correlation Cronbach’s alpha Aggregated into:

Five questions about fit P = 0.000 0.909 Perceived fit

Four questions about fan identification P = 0.000 0.920 Perceived fan identification

Six questions about team performance P < 0.05 0.813 Perceived team performance

Furthermore, by testing the questions about the two moderating variables, analysis revealed that the outcomes of the perceived team performance and the perceived fan identification questions correlate with each other. Analysis revealed a Pearson correlation of p=0.000 and a Cronbach’s alpha of 0.856. This indicates multicollinearity, which means that both variables are highly correlated to each other, and may provide redundant information about the outcomes. So in this research, the variable perceived team performance affects the variable perceived fan identification. This may be two-sided, so when the perceived team performance is high, the perceived fan identification will also be high. However, as FC Groningen underperformed this season, it is likely that the perceived fan identification is therefore also lower.

In table 11, the mean scores and the standard deviations of perceived fit, perceived fan identification and perceived team performance can be found. Remarkeble are the outcomes of the perceived team performance (M: 3.12) and perceived fan identification (M: 5.75). When relating these outcomes to the multicollinearity of these two variables, these values are assumed to be even higher when FC Groningen is playing a good season.

1

25 Table 11: Means and standard deviations of the moderating and independent variables (seven-point scale)

Mean SD Variance

Perceived fit 4,95 1,33 1,76

Perceived team performance 3,12 0,96 0,92

Perceived fan identification 5,75 1,43 2,04

Another question in the survey asked whether the respondents were season ticket holder of FC Groningen during the current season. In order to check if there were differences between the two groups (season ticket holders and non season ticket holders) and the moderating variable fan identification, the data was split and the descriptives of both variables were performed. Analysis revealed that the group which consist of season ticket holders have a higher level of fan identification (M: 6.22, SD: 0.74) than the non season ticket holders (M: 5.43, SD: 1.48).

4.2 Empirical results

4.2.1 Outcomes of consumers’ sponsors recall

The outcome of the unaided recall was 0.939 which indicates that almost 94% of the respondents knew the main sponsor of FC Groningen during the 2011/2012 season. Therefore this can be seen as a very good score.

In order to see how the respondents scored on the aided recall question a special equation was prepared (this equation can be found in section 3.3). When the equation was used, de formula becomes: (3.27-2.1)/0.0696 = 16.81. The unaided recall is therefore 16.81. This outcome is approximately standard normally distributed. As the guess probability for this research was 2.1 and the outcome of the aided recall is 16.81 it can be said that the respondents performed very well on this question as the aided recall score is much higher than the guess probability.

To check the respondents’ ability to select the right sponsors out of a list of ten firms, the Signal Detection Analysis of Green and Swets (1996) was used. As stated in the method section of the dependent variable, the respondents’ sensitive scores (A’) was measured. For this research, the performance of each respondent can be summarized in either a hit rate or a false alarm rate. The identification results can be found in table 12.

Table 12: The different types of responses of the Signal Detection Analysis

Identification results Percentage Hit rate 77,65%

False Alarm rate 18,70%

26 When further looking at the outcomes of the aided recall question, referring to table 13, it can be said that the respondents on average choose 4.2 firms out of the list. On average the respondents choose 3.27 correct sponsors from the list of firms. In table 14 is shown how many times each company was chosen and the percentages of the aided recall question.

A remarkable result here is that almost half of the respondents (47%) thought that the company ‘’Greenchoice’’ is a sponsor of FC Groningen, while in fact they are not. A possibility here might be, is that the name ‘’Greenchoice’’ refers to ‘’green’’ which is actually what Essent and FC Groningen tries to communicate through their sponsor relationship. This may be a reason that so many respondents picked the company as a sponsor.

A second remarkable result of the aided recall question is that only 13.9% of the respondents picked ‘’Hummel Tuintotaal’’ as a sponsor, whereas their competitor ‘’Kamphuis Hoveniers’’ which is not a sponsor of FC Groningen is selected by 12.5% of the respondents. Overall it can be said that 98.6 percent of the respondents were able to correctly recall at least one sponsor, while 53.5 percent correctly recalled three or more sponsors. Furthermore, looking at the results of the recall of Essent, it can be said that the aided recall of Essent is even higher when comparing it to the unaided recall of Essent. This might have to deal with the first question which asks who the main sponsor of FC Groningen is. Most respondents (94%) already knew that this was Essent. However, by also showing Essent in the aided recall question, it might be that respondents, who were not sure whether Essent was the main sponsor, now identified the company as a sponsor of FC Groningen.

Table 13: Results of the aided recall question (on a scale of 0-5)

Answers of the respondents

Average number of firms chosen 4,21

Number of correct answers 3,27

Table 14: Results of the aided recall question + the durations of the various sponsor co-operations The companies with a * are official sponsors of FC Groningen during the season 2011/2012

Number of times chosen by resp. Percentage Sponsor since:

Essent * 212 98,6% 01-07-2011

Greenchoice 102 47,4%

Bos Incasso * 202 96,9% At least since 01-01-20062

WVM Deurwaarders 43 20%

Hummel Tuintotaal * 30 13,9% 01-07-2006

Kamphuis Hoveniers 27 12,5%

La Compagnie * 89 41,4% At least since 01-01-2006

Storm MC 16 7,4%

Gasterra * 169 78,6% 01-07-2011

Shell 15 7,0%

One explanation for the differences in sponsor recall might be the duration of the sponsorship. Therefore the starting dates of the various sponsor co-operations are also mentioned in table eleven. When looking at these durations, it can be seen that Bos Incasso and La Compagnie have the longest duration from these five sponsors.

2

27 The recall score of La Compagnie is quite good when comparing it to the non-sponsor Storm MC. The recall score of Bos Incasso is almost one hundred percent, but this will relate to the fact that they have been a main sponsor of FC Groningen for some years. The scores of Gasterra and Hummel Tuintoaal are less easy to explain. Gasterra is together with Essent, one of FC Groningen’s ‘’green partners’’ and will probably enjoy the prominence of the green message between Essent and FC Groningen. Hummel Tuintotaal however is already a sponsor since 2006 but their recall can be considered as very low.

4.2.2 Additional testing

When looking at the responses of the two questions which were asked on behalf of Essent, it can be said that 158 respondents (73,5% of the sample) filled in the question which was about selecting keywords from a given list. The percentages of the keywords which were chosen by the respondents and also had a score above five percent can be found in table 15.

Table 15: Percentages of the keywords chosen by respondents Items % chosen by respondents

Groen 56,7% Duurzaam 45,1% Doordacht 32,1% Passend 27,9% Geschikt 27,4% Slim 24,6% Begrijpelijk 21,4% Mooi 17,7% Logisch 17,2% Duidelijk 15,8% Kloppend 14,4% Relevant 13,5%

When looking at table 15, it can be said that (almost) half of the respondents think of the relationship between Essent and FC Groningen as ‘green’ and ‘sustainable’. Other high-scoring keywords are ‘well fitting’ which scores 27.9% of the respondents, ‘appropriate’ with a score of 27.4% , ‘smart’ which scores 24.6% and ‘understandable’ with a score of 21.4%. When looking at the keywords which gained a score less than five percent, the two most striking keywords are ‘hypocrite’ (4,2%) and ‘fake’ (3,2%).

In appendix IV, the responses to the other question about Essent can be found (‘’What do you think, is the overall message of the cooperation between FC Groningen and Essent?’’). In total, 68.8 percent of the respondents indicated some of the following items: green, green energy, sustainability. Furthermore, ten percent of the respondents exactly knew the slogan of FC Groningen and Essent, namely together completely green (‘’samen helemaal groen’’). These results indicate that the respondents have a pretty good idea what Essent and FC Groningen try to communicate by their cooperation.

28 Table 16: Results of the additional questions about Essent (seven-point scale)

Mean SD Variance

Generall impression about Essent 4,92 1,42 2,02

Consumers preference concerning Essent products 4,19 1,71 2,95

When looking at the frequencies of the outcomes on the independent and the moderating variables, it can be said that almost eighty percent of the respondents (78%) gives a score between one and four on the variable perceived team performance. Fifty percent of the respondent gave a score of six of higher on the variable perceived fan identification. And almost fifty percent (47%) gave a score of five or higher on the variable perceived fit. These scores are all on a scale of one to seven, so the score on perceived team performance can be considered as very low.

This is likely to be related to the performance of the team. However, the score on fan identification can be seen as a very positive one. And also the score on the perceived fit can be considered as positive for Essent and FC Groningen.

4.2.3 Measuring the variation of coefficients

In order to test the distribution with respect to the averages of the independent variable and the moderating variables, the variation of coefficients is measured. The following formula has been used for this purpose:

In table 17, the calculated coefficient variables of the independent variable and the moderating variables can be found. When looking at the results of the coefficient variation, it can be said that the variation of the coefficients of the measured variables lies between 20 and 30 percent. According to the guidelines of the Microdata User Guide (2006), see table 18, these coefficients can therefore been seen as having a marginal quality level of estimate.

Table 17: Coefficient variations of the IV and MV’s (seven-point scale)

Mean SD Coefficient variation

Aided recall 0.94 0.24 25.35%

Perceived fit 4.95 1.33 26.87%

Perceived fan identification 5.75 1.33 23.13%

Perceived team performance 3.12 0.96 30.77%

Table 18: Guidelines of the coefficients of variation (cited by the Microdata User Guide, 2006) Quality level of estimate Guidelines

1) Acceptable Estimates have a sample size of 30 or more, and low coefficients of variation in the range of 0.0% to 16.5%

2) Marginal Estimates have a sample size of 30 or more, and high coefficients of variation in the range of 16.6% to 33.3%.

29

4.3 Hypotheses testing

In order the check the first hypothesis of this research, does fit between a sponsor and a club has an effect on the sponsor recall of consumers, a regression analysis was performed on a 95% reliability level. In this model the outcome of the sensitivity score (A’) (aided recall) was used as the dependent variable, while the aggregated variable perceived fit was used as the independent variable. This recall variable is chosen as the unaided recall is already 0.939 within the sample which means that this variable is unlikely to become even higher through the moderating variables. In table 19, the results of this regression model can be found. This regression however, was non-significant, R= .036, F (1,156) = .202, p = .654. So it can be said that the perceived fit between a sponsor and a sports club does not affect the sponsor recall of consumers, B = 0.004, p =.654.

Table 19: Results of the regression analysis, H1 β Sig.

Overall regression .654

Perceived fit 0.004 .654

To test the second hypothesis of this research, again a regression analysis has been performed on a 95% reliability level. Here, the dependent variable was again the outcome of the sensitivity score (aided recall). The moderating variables perceived fan identification and perceived team performance were used as the independent variables in this research. This regression was also non-significant (see table 20), R= .089, F (4,153) = .308, p = .972. So the perceived fan identification does not lead to a higher sponsor recall of consumers, B = -0.005, p =.852. Neither does the perceived team performance have a significant effect on consumers’ sponsor recall, B = -0.014, p =.799.

Also the cross products of the perceived team performance and perceived fan identification have been measured within this regression, these variable were, logically, also not significant. The cross product of perceived fan identification measured: B = 0.001, p =.819. And the cross product of perceived team performance measured: B = 0.001, p =.956.

Table 20: Results of the regression analysis, H2

β Sig.

Perceived Fan Identification -0.005 .852 Perceived Team Performance -0.014 .799 Cross product Fan Identification 0.001 .819 Cross product Team Performance 0.001 .956