Faculty of Social and Behavioural Science
Communication Science
Corporate Social Responsibility
in European professional football
A comparison between Bundesliga and Premier League
from a stakeholder perspective
Master’s Thesis
Graduate School of Communication Master’s programme Communication Science
Corporate Communication
Submitted by Lukas Böhm Student number 12843695
Supervisor Dr. Pytrik Schafraad Date of Submission 26/06/2020
Abstract
Corporate Social Responsibility is playing an increasingly important role in the football business. Since football clubs are no longer perceived merely as sports clubs, but rather as economic corporations, societal expectations are rising. Furthermore, football clubs are considered as stakeholder organizations with a strong embeddedness in their communities. Therefore, they rely on strong relationships and thus must embrace CSR strategies. Even though CSR is highly connected to stakeholder theory in sporting contexts in research, fans as key stakeholders are often not included in empirical studies.
This master thesis examines Corporate Social Responsibility within European professional football from a stakeholder perspective and compares the two top leagues Premier League and Bundesliga. The research topic is investigated by applying a two-study research design: (1) A content analysis is conducted to examine the clubs’ CSR engagement and its accompanying communication strategies. In addition, (2) an online survey investigates fans’ perception of CSR and its impact on the reputation of football clubs under consideration of additional factors such as team identification and perceived dialogue possibilities. Based on this, the thesis subsequently analyzes how those perceptions align with the football clubs’ disclosed engagement and its communication strategies and gives first insights on how a 'CSR stakeholder-company fit' approach could be implemented in a sporting context.
Although Bundesliga and Premier League clubs show only small differences in the type of CSR engagement and communication strategies, the clubs are dealing much more intensively with the topic of CSR than they did a few years ago. Furthermore, the findings on the impact of CSR indicate a strong significant positive effect of CSR perception on corporate reputation, though not differing per league. Team identification plays no significant role in this
relationship, while the degree to which fans perceive dialogue possibilities regarding CSR has an indirect effect on corporate reputation. The analysis regarding the alignment between actual CSR engagement and communication with fans’ perceptions reveal discrepancies. Although clubs engage in some CSR themes and dialogue strategies intensively, they are not as strongly perceived by their fans. Based on these findings, implications for sports clubs and recommendations for future research are provided.
3
Introduction
In early 2020, the coronavirus spread rapidly and developed into a worldwide pandemic. National economies and health systems were facing one of the worst crises in recent history. As a result, many governments were forced to implement emergency programs to protect corporations and employees. The British government for example introduced financial state aid for employees who were sent on short time by their companies. However, one topic seemed to receive particular attention among all the negative headlines for some time in the UK: the attempt of Liverpool FC to benefit from those federal aid payments by sending parts of their employees on furlough. It caused a lot of discussion and controversy within society and on social media since the company was in a good financial position. The well-known newspaper 'The Telegraph' headlined: "Liverpool FC were supposed to be better than this – their use of government scheme feels like a betrayal" (Edwards, 2020). The reactions from society to the way the club dealt with its own employees were enormous and led so far that Liverpool finally backed down and apologized for its actions.
The example of Liverpool FC shows that football clubs in professional sports have long since ceased to be considered merely as sports clubs, but rather as economic corporations, with an increased responsibility in society. Additionally, the reaction of the football club demonstrates the enormous influence of societal expectations on companies regarding their corporate citizenship. Viewing this from a corporate communications perspective shows which severe impacts wrong corporate decisions can have on the reputation of a company and demonstrates the importance of integrating stakeholders in decisions on issues affecting society. To meet those expectations, companies engage to an ever-increasing extent in Corporate Social Responsibility (CSR) (e.g. Carroll, 2015).
Even though CSR is nothing new in professional sports, it plays an increasingly important role. Since football clubs are considered as stakeholder organizations with a strong
4 must embrace CSR strategies for philanthropic reasons (Hamil & Morrow, 2011). Although CSR is highly connected to stakeholder theory in a sporting context, fans as key stakeholders are often not included in empirical research (Fifka & Jaeger, 2020). Some studies point out that especially the impact of CSR activities in connection with stakeholders of sport clubs has still not been adequately addressed and needs further investigation (e.g. Walker & Kent, 2009).
This master thesis aims at filling parts of this research gap and investigates the impact of clubs’ CSR engagement and its accompanying communication strategies in two top European football leagues by including the fans’ perception. A comparison between the selected leagues – namely the Bundesliga and the Premier League – is highly interesting due to the difference in their orientation regarding their values. Whereas the German Bundesliga clubs seem to be far more fan-oriented with a strong relationship to its communities, which is shown by the fact that the Bundesliga is the best supported league by attendance (Besson, Poli & Ravenel, 2019), the English Premiere League seems rather profit-oriented. With a revenue generation of 6 billion dollars in 2017/18 it is 72% larger regarding revenue volume than the nearest competitor, the Bundesliga (Kidd, 2019). Therefore, football clubs’ CSR strategies might differ depending on the league in which the clubs compete.
To address the described research gaps, this master thesis investigates Corporate Social Responsibility within professional football and compares the Bundesliga and the Premier League regarding their CSR engagement and its accompanying communication. It especially focuses on fans’ perception of CSR, its impact on the reputation of a football club and how those perceptions align with the football clubs’ disclosed engagement and its communication. Therefore, the following research questions will be examined:
5 RQ: Does the engagement within Corporate Social Responsibility in European professional
football differ between Bundesliga and Premier League clubs? Is there a difference in the perception of CSR activities and its impact on corporate reputation between the supporters of both leagues? Do engagement and perceptions align with each other?
Theoretical Framework
Corporate Social Responsibility
Nowadays, companies must adapt constantly to fast changing environments. Additionally, organizations are required to position themselves on many socially highly discussed topics and show their commitment on social and environmental issues to further meet growing expectations of diverse stakeholder groups (Basu & Palazzo, 2008; Young & Marais, 2012). As a result, more and more organizations focus on Corporate Social Responsibility (CSR) as this portraits "[...] the responsibility of enterprises for their impact on society" (European Commission, 2019).
The topic of CSR has concerned both economy and research for the last three decades (Athanasopoulou, Douvis & Kyriakis, 2011). Especially over the last 10 years, an increase in CSR engagement can be observed not only in economy but also in science. One reason for growing CSR efforts on company side was the introduction of CSR reporting standards like the Global Reporting Initiative (GRI). This made the concept of CSR more measurable and led to a further increase in numbers of companies that were involved in CSR (Nikolaeva & Bicho, 2011; Vartiak, 2016). The growth in companies engaging in CSR made the issue also increasingly important for research from various viewpoints.
In general, giving one universal definition of Corporate Social Responsibility from a research perspective is somewhat impossible as numerous explanations exist (Carroll, 2015). Considering this fact, this thesis will focus on a basic and in research widely used definition by Carroll (1991, 2015). It aligns with academic consensus viewing CSR as a construct that
6 represents a set of actions which seems to promote a social good, goes beyond the explicit financial interests of a company and is not required by law (McWilliams & Siegel, 2000). Carroll (2015) describes CSR as a general term which argues that company managers should carefully consider the organization’s impacts on society. The author continues that it includes those actions which executives and companies take to protect and improve the welfare of society along with business’s own interests. ‘Protection’ in Carroll’s definition means avoiding negative impacts like pollution, deforestation or disrespecting legal regulations. ‘Improving’, on the other hand, aims at creating positive benefits for society including for example philanthropy or community development. As a basis for this, the author gives another broad definition for CSR in a paper in the early 90’s which is amongst the most used definitions in CSR literature. He builds a simple framework including economical, legal, ethical and philanthropic factors being important for an organization regarding CSR (Carroll, 1991). With his “pyramid of CSR” the author claims that a company must first fulfill
economic responsibilities – simply being profitable – to satisfy society’s other expectations. Furthermore, another inevitable basis for being socially responsible is meeting all legal regulations. Only then, companies are able to act ethically and philanthropically responsible. Therefore, these two factors occupy the uppermost parts of the four-level pyramid (Carroll, 1991).
The perspectives from which CSR is investigated in research are manifold. Existing literature focuses on either conceptual/theoretical work – dominating all other viewpoints –, motives-oriented work or outcome-oriented work and puts emphasis on various sectors (Walker & Kent, 2009). However, many of the studies agree on one consensus: the important role of stakeholders of organizations regarding Corporate Social Responsibility, as CSR is built on their expectations. From a theoretical political-normative view on CSR, companies must achieve cognitive and moral legitimacy through Habermasian consensus-building with their stakeholders (Scherer & Palazzo, 2011). While cognitive legitimacy is given when
7 companies adapt to community values, moral legitimacy refers to moral judgements on
organizations which can only be reached through a communicative deliberative process (Scherer & Palazzo, 2011; Schultz, Castelló & Morsing, 2013). Motives-oriented work shows, that executives use CSR amongst others mainly to satisfy their stakeholders (Walters & Tacon, 2011) and from an economic perspective, research literature often focuses on investigating the extrinsic impact of CSR on consumer groups (e.g. McWilliams & Siegel, 2000; Walker & Kent, 2009; Öberseder, Schlegelmilch & Gruber, 2011). According to Du, Bhattacharya and Sen (2010) those economic impacts of being a good corporate citizen can reach from increased advocacy behavior to a rise in consumer loyalty and overall profits.
Corporate Social Responsibility in Professional Football
One sector for which stakeholders are particularly important is professional sports. While Corporate Social Responsibility has long been discussed in research, it has also gained increasing importance in European sports within the last years – especially in football as it is by far the most popular sport on this continent (Van Bottenburg, 2011; Fifka & Jaeger, 2020).
Although football clubs are often focusing on philanthropic reasons for CSR to further attract their stakeholders, professional sport clubs are also business organizations which aim at improving their profits strategically. Therefore, CSR strategies are applied due to pragmatic respectively extrinsic reasons as CSR can also be a strategic tool for reaching specific
beneficial outcomes (Du et al., 2010; Anagnostopoulos, Byers, & Kolyperas, 2017).
Additionally, football inevitably experiences the same pressures as other sectors of industry due to commercialization and is therefore strongly engaging in CSR (Hovemann, Breitbarth & Walzel, 2011).
Various research perspectives exist in the context of CSR communication and professional sports, though often focusing on US sports like American football, ice hockey or basketball (e.g. Babiak & Wolfe, 2009; Walker & Kent, 2009; Heinze, Soderstrom & Zdroik, 2014;
8
Lacey & Kennett-Hensel, 2016). Additionally, regarding the fact that sports clubs are
considered as stakeholder organizations, it is not surprising that CSR in a professional sports context in connection with stakeholder theory has heavily evolved in academia (Fifka & Jaeger, 2020). The authors Fifka and Jaeger (2020, p. 62), however, state that “these results [from US sports] are not necessarily applicable for the case of European football, since
political and legal frameworks and the sports system itself are very heterogeneous”. Although various studies exist focusing on the case of European sports and CSR – especially on
football, some areas are not covered satisfyingly. Existing research on the topic includes not only work investigating reasons and motivations for implementing a CSR strategy through qualitative interviews with CSR executives (e.g. Walters & Tacon, 2011; Kolyperas, Morrow & Sparks, 2015), but also content analyses on CSR issues addressed by football clubs (e.g. Hovemann et al., 2011; Kolyperas 2012) or clubs’ CSR communication (e.g. François, Bayle & Gond, 2019). The impact of CSR in football, however, was investigated just in a few studies and if so, not always related to Europe. A study by Liu, Wilson, Plumley and Chen (2019) for instance examines the influence of perceived CSR on fan-based patronage intentions in Chinese professional football, while another study focuses on two clubs in France, investigating CSR engagement effects on customer-based brand equity (Blumrodt, Bryson & Flanagan, 2012).
The content analyses of Hovemann et al. (2011) and Kolyperas (2012) showed that there are generally differences in CSR communication between different football clubs (25 largest European football clubs) and leagues (Switzerland, Germany and England). While the football clubs “[…] differ in the way they perceive, adopt, execute and communicate the concept [of CSR]” (Kolyperas, 2012, p. 123), the leagues vary concerning different levels of CSR commitment, communication and formal integration, with England leading the field of CSR (Hovemann et al., 2011). However, since the analyses were carried out several years ago and the topic has developed enormously within the last decade, it is necessary to refresh the
9 results of the studies for the Bundesliga and the Premier League. Due to the lack of actuality, the studies do not provide a basis for an adequate comparison between perceived CSR by football fans and implemented CSR by football clubs nowadays.
Also, as this research aims at investigating CSR and football from a stakeholder
perspective, a new part will be added, namely stakeholder involvement regarding clubs’ CSR communication. Empirical research has not put an emphasis on this topic so far, although several research papers address its importance within CSR communication (e.g. Greenwood, 2007; O’Riordan & Fairbrass, 2008; O’Riordan & Fairbrass 2014; Fifka & Jaeger, 2020). According to Greenwood (2007) involvement strategies can be particularly used as an instrument to achieve various company objectives including consent, control or accountability.
On this basis and due to the lack of actuality of previous content analyses, the following two sub-research questions need to be examined on a league basis:
Sub-Research Question 1:
Which CSR issues are addressed by football clubs from the Premier League and the Bundesliga?
Sub-Research Question 2:
Which channels do football clubs use to communicate their CSR activities? Do they use certain involvement strategies for integrating their fans when communicating about CSR?
Impact of CSR Communication on Corporate Reputation
Different effects of Corporate Social Responsibility on consumers and, in turn, on the company have been investigated and confirmed not only in a business-related context, but also in sports environment. Especially the effects of CSR on corporate reputation have been examined since improving a company’s image is one of the main goals when engaging in CSR. Carroll (2016, p. 4) states that “[…] most companies engage in philanthropy as a
10 reputation […].”. Additionally, reputation literature in general suggests that a company's overall reputation is better the more it contributes to society (e.g. Fombrun & Shanley, 1990). Empirical research has been able to confirm this positive relationship (e.g. Gatti, Caruana & Snehota, 2012; Bianchi, Bruno & Sarabia-Sanchez, 2019). A positive corporate reputation is particularly important for sports clubs, as they are highly dependent on their supporters and connected to their community. The positive effect of (perceived) CSR on corporate reputation has also been proven in the area of professional sports businesses (e.g. Walker & Kent, 2009; Lacey & Kennett-Hensel, 2016). However, since the effect has been validated in American sports clubs of the NFL/NBA and therefore, as mentioned before, cannot simply be
transferred to European football, this relationship will be examined for the German Bundesliga and the English Premier League. Furthermore, on the basis of both, the earlier found differences regarding CSR efforts of Bundesliga and Premier League clubs by Kolyperas (2012) as well as the fact that the Bundesliga is said to be the most supported league worldwide – confirming strong fan relationships – (Besson et al., 2019), it will be exploratively examined if the effect of perceived CSR on club reputation will differ per league. Based on the results of Walker and Kent (2009), the following hypothesis is thus formulated:
H1: The degree to which fans perceive a football club’s CSR positively influences the reputation of a club.
H1a: This effect will be stronger for Bundesliga fans than for supporters of the Premier League.
The Importance of Team Identification
In their integrative framework of CSR in professional football, Fifka and Jaeger (2020) recognize fans and club members as one of the key areas of CSR in sport. Adding to that, Jung (2012) notes that the financial basement of every sports club is highly dependent on community support. One variable, which is often included when researching CSR in sports in
11 connection with fans, is the effect of team identification. It can be described as “[…] an
individual’s level of support for a team that includes a personal acknowledgment of his or her emotional attachment to the team” (Jung, 2012, p. 4). One reason for including team
identification is, that highly identified fans often see their team as a reflection of themselves and tend to have a high difference in reactions compared to those who are low in
identification with a team (Jung, 2012). Individuals with a high club identification are more likely to interact with the club for instance in terms of visiting the stadium or purchasing merchandise (Jung, 2012).
In relation to CSR, Walker and Kent (2009) found out, that fans of two NFL clubs who strongly identified themselves with the team, were less affected by CSR initiatives in the way how they perceived the reputation of their supported club than those with low team
identification. This moderating effect will also be tested for the first hypothesis of this study to prove the investigated outcome by Walker and Kent (2009) and whether this can also be observed for Bundesliga and Premier League football fans. On this basis, the following second hypothesis will be proposed:
H2: Team identification moderates the positive effect of CSR perception on corporate reputation negatively in such a way, that the effect will be less strong for high identified fans than for those with a low team identification.
Based on the differences in fan cultures and the associated expectation of stronger fan relationships in the Bundesliga, the following hypothesis H2a will be derived:
H2a: This moderation effect will be stronger for Bundesliga fans than for supporters of the Premier League.
12
Stakeholder Dialogue Strategies
The previous sections have highlighted the importance of fans as key stakeholders of football clubs. When connecting CSR with stakeholder theory, research also shows the significance of involving stakeholder expectations and interests concerning corporate decisions with continuous stakeholder dialogue (O’Riordan & Fairbrass, 2014; Fifka & Jaeger, 2020). Increased interaction with key stakeholders is important and promotes the development of knowledge and expertise on specific issues faced by companies (Maon, Lindgreen & Swaen, 2009). Stakeholder involvement strategies in CSR are based on two-way symmetric communication, meaning stakeholders are integrated in decision-making processes and co-construct corporate CSR efforts through constant pro-active dialogue between
stakeholders and companies (Morsing & Schultz, 2006). Actions for implementing such strategies can vary and include for instance open dialogue on different social media platforms, participative websites or round tables with important stakeholders (Monfort, Villagra & López-Vázquez, 2019). In their conceptual framework of managing CSR stakeholder engagement the authors O’Riordan and Fairbrass (2014) highlight, that stakeholder
engagement within CSR communication can impact a firm’s credibility as well as the corporate image in society. However, they also point out that empirical studies must be carried out to verify the effects of stakeholder engagement.
Dialogical respectively involvement strategies for football clubs’ CSR activities have not been investigated by empirical studies before. Consequently, the following hypotheses are examined exploratively on basis of the literature for both the Premier League and the Bundesliga to point out a first possible direction for further research and to give initial recommendations for football and sports clubs in general:
H3: The degree of fan perceptions of dialogue possibilities regarding club CSR has a positive impact on corporate reputation.
13 H4: The degree of fan perceptions of dialogue possibilities regarding club CSR has a
positive impact on CSR perception.
Based on Hypothesis 1, 3 and 4 an indirect relationship between the perception of dialogue possibilities and corporate reputation mediated by the perception of CSR will be expected:
H5: The perception of CSR mediates the effect between perceived dialogue strategies and corporate reputation positively.
CSR Company-Stakeholder Fit
Academia does not only emphasize the importance of CSR or accompanying dialogue strategies in order to achieve benefits for a corporation, but also the significance of an alignment of CSR and communication with consumer expectations and perceptions. This ‘CSR company-stakeholder fit’ is especially important as competitiveness is nowadays highly dependent on the relationship between a company and its stakeholders (Calabrese, Costa, Menichini & Rosati, 2012). Menichini and Rosati (2014) state that business returns from CSR activities, like corporate reputation or company-stakeholder identity, are highly dependent on how stakeholders perceive the company’s CSR engagement. In turn, for companies it is necessary to understand stakeholder feedback and constantly align to their expectations and perceptions. If a company's CSR activities and policies do not match the actual perception of stakeholders, the positive impact of CSR could be diminished (Becker-Olsen, Cudmore & Hill, 2006). Therefore, it is essential to measure the congruence or gap between the demands and perceptions of stakeholders and the actual commitment of the company (Calabrese et al., 2012; Menichini & Rosati, 2014).
As the importance of stakeholders for sports clubs was underlined before, the CSR
company-stakeholder fit is especially significant for football clubs. Following this fact and the two first sub-research questions as well as the five hypotheses, a final third sub-research question needs to be examined:
14 Sub-Research Question 3:
Is there an alignment of the clubs’ disclosed CSR engagement and its accompanying dialogue strategies with the fans’ perceptions of it?
Conceptual Model
In the theoretical part of this thesis, five hypotheses were developed based on previous literature to investigate the effects of CSR in football, including stakeholder engagement. For a better overview of the construct, the following conceptual model is developed:
Figure 1: Conceptual Model
Research design
The aim of this study is the investigation of the CSR engagement and communication of Bundesliga and Premier League football clubs. Additionally, the influence of football clubs‘ Corporate Social Responsibility perceived by fans on the clubs’ reputation will be examined. Other factors such as team identity and the perceived possibility for entering dialogue with the clubs will also be considered as shown in the before mentioned conceptual model.
15 Bundesliga and the English Premier League. Finally, it will be examined if there is an
alignment between the clubs’ actual CSR engagement and communication with the perception by the fans. In order to measure the alignment between disclosed CSR commitment and how consumers (fans) perceive it, the authors Menichini and Rosati (2014) suggest a two-way approach with both, a content analysis of a company’s CSR efforts and consumer surveys on their perceptions.
Therefore, to answer the (sub-)research questions and the hypotheses, two research studies had to be conducted: a content analysis on the clubs’ CSR engagement and an online survey on the fans’ perceptions. Both procedures will be explained in detail in the following section.
Study 1: Football Clubs’ Communication about CSR
The first study was a quantitative content analysis which was conducted on the websites and CSR reports of 24 Bundesliga and Premier League football clubs regarding their CSR communication. The data was gathered to get a better understanding of football clubs’ current CSR engagement and their related communication strategies. The reason for choosing the websites as object of investigation is that they can be seen as a company’s reliable main communication platform on which all topics concerning the organization are addressed. Other studies which investigated CSR with the help of content analyses proof this as a legitimate decision (e.g. Pollach, 2005; Kolyperas, 2012).
Selection of Research Units
Since the results of the analysis of the clubs’ CSR content should be considered in connection with the results of the online survey and be also representative for the leagues, it was important to select various clubs from both examined leagues. In principle, all 38 clubs of the two leagues are eligible for the analysis, but since the research was conducted by just one researcher, the number of clubs under consideration had to be limited.
16 Before starting the online survey as the second part of the research, it was assumed that in general fans of the more popular clubs would be more likely to be reached by the survey. Therefore, the more famous clubs were selected for the content analysis. In total, 24 football clubs (12 per league) were chosen based on club size, measured by their annual turnover in euro (Bundesliga) and in dollar (Premier League) (Gough, 2019; Sky Sport, 2018). For the Bundesliga, clubs like FC Bayern München, Borussia Dortmund or SV Werder Bremen were chosen, while for the Premier League, clubs like Arsenal FC, Liverpool FC or West Ham United were considered for the analysis.1 The examined period for which the CSR content was analyzed included the last football season 18/19 and the current season 19/20, as some CSR communication is about past activities and CSR or annual reports are often only available for last seasons.
Objects of investigation were the clubs’ official internet sites (e.g. CSR chapter, press release section), available annual and CSR reports from latest 2018 as well as websites of club foundations and official external webpages.
Coding Instrument and Pretest
Basically, the codebook serves to operationalize the variables that are necessary to analyze selected characteristics of a text. It is the basic guide for a textual content analysis for the researcher and can be executed in different ways. In this case, Qualtrics – an online platform for surveys – was used to carry out the analysis in order to have all the data available in digital form. The codebook included some basic coder-instructions at the beginning as well as
various keywords derived from Kolyperas (2012). They were used to search the website in case no obvious CSR sections could have been found. CSR activities are often described with
17 different wordings - key terms like “foundation”, “sustainability” or “corporate citizenship” represent the construct as well.2
In the first stage, the codebook comprised a total of nine content variables. However, to be able to use these categories for research in an optimal and reliable way, they had to be
checked and adjusted by conducting a pretest before the actual content analysis. For this purpose, two selected websites of one German and one English football club were coded by the researcher and his supervisor. This test sample size corresponds approximately to the research pretest standard of 10% of the total sample. As a result, it turned out that the
codebook was basically suitable for the analysis. In the end, only a few adjustments had to be made. Two variables were not considered necessary for this study and were therefore deleted from the codebook. Other changes included minor modifications in the coding instructions and adjustments for items of two variables. Finally, the codebook included 9 descriptive categories which will be explained in more depth within the following part.
Operationalization of the Variables and Reliability
Several content analyses which analyzed Corporate Social Responsibility in the context of sport were found during the literature research. The variables from these previous studies provided the basis for most of the variables in this content analysis, since they were created primarily deductive and were therefore already validated.3 Additionally, open text fields were included for the most categories in order to inductively identify new issues not acknowledged by previous research. All variables served the mere descriptive evaluation to identify the disclosure of CSR themes, communication channels, targeted stakeholders and the use of different dialogue strategies of football clubs from both leagues. Therefore, all of them were coded as dichotomous latent variables (existent vs. not existent). After coding the club’s name
2 For further information on the key terms see the codebook in Appendix A. 3 For further information on all variables see the codebook in Appendix A.
18 and league it competes in, the different types of general CSR communication mediums used by the club were analyzed.
The items for one of the main variables, namely “CSR themes” are based on findings from the authors Kolyperas (2012) and Kolyperas et al. (2015) who researched CSR within the professional football business. The variable contained for example headings for CSR activities like community football, charity/fundraising activities, fan-led initiatives or environment and sustainability. An open text field was used to code also new topics, which had not been addressed by previous research. The variable which aimed at researching the usage of different involvement respectively dialogical strategies is based on definitions of Morsing and Schultz (2006) as well as Colleoni (2013), who researched stakeholder
involvement strategies regarding CSR communication. The operationalization of the variable is derived from suggestions of a study exploring stakeholders’ dialogue and CSR on Twitter
(Monfort et al., 2019). The variable encoded whether CSR communication of football clubs was accompanied by different stakeholder involvement strategies which can be integrated or communicated on a website. Those strategies could either include links to CSR postings on social media and debating forums or simply reports on meetings with fans (round tables) regarding CSR topics.
To verify the reliability of the codebook as a measurement tool the intra-coder reliability had to be calculated. For this purpose, the percentage of agreement per item for all coded variables was calculated. This coefficient is suitable due to the small sample size and the fact that just one person coded the sample. For this purpose, four websites, representing almost 20% of the sample size, were coded again and then compared with the originally recorded values. The overall percentage of agreement was 94% (ranging from 92% to 100%) for the reliability testing. It can be interpreted as a good value, however, it could also be higher for this sample size. The open text fields of some variables made an exact agreement in some cases difficult, which explains a slightly worse percentage of agreement for some categories.
19
Study 2: Fan Perceptions of Club CSR
The second study aimed at investigating the influence of football clubs’ CSR activities perceived by fans on club reputation, considering the impact of team identification and dialogue strategies. This research topic was investigated with a quantitative online survey. The method was best for reaching the target group, namely fans of the Premier League and the Bundesliga, as it gives access to unique populations (Wright, 2005).
Survey Design and Pretest
In order to test the hypotheses an approximately 7-minute cross-sectional online survey was conducted over a period of two weeks. The questionnaire was self-reported asking the participants for their opinion on the research topic. The chosen time period can be explained by the short period in which the study was conducted (13 weeks). The modality of the survey design was an online questionnaire with using the platform Qualtrics. One major benefit of an online survey is the speed with which the respondents can be interviewed in order to get a large sample in a relatively short time period and with a minimum of financial costs (Wright, 2005). In addition, this survey format provides the respondents with the opportunity to answer the questions on any digital device. This increased flexibility can in turn lead to better
response rates. However, the response rate is also one of the major drawbacks of using an online survey, because potential participants could feel less motivated to complete the survey if they are not contacted personally. The possibility of a sample distortion due to multiple answers was avoided from the beginning by survey settings, which only allowed to use the anonymous link once. Another disadvantage is the fact that participants have no option for questions regarding possible misunderstandings during the survey. In order to avoid the latter disadvantage, a pretest with approximately 10% of the sample size (N=24) was carried out before the start of the actual study.
20 The participants of the pretest were German- and English-speaking users in order to get feedback from both target groups. One major suggestion within the feedback was related to the language of the survey. Originally, the questionnaire was designed in English, however, four German pretest-participants responded, that they found it difficult to distinguish between certain concepts due to the language barrier. As a result, to avoid the possibility of reaching a too low sample size of Bundesliga fans, the survey was translated into German. Other
feedback from the test version included improvements in some explanations as well as small suggestions for scale enhancement and one additional control question. The translation into another language led to a second pretest of the survey with only German participants for confirmation of those translations. The second pretest (N=5) did not reveal any significant changes.
Finally, the survey was available in German and English and comprised three different sections. The first part of the questionnaire included an introduction, an agreement of participation and more general questions like the participants preferred football league, their favorite club and their football fandom. The latter construct was derived from Wann (2002) and Absten (2011) and was used as a control variable during the analysis. The main part contained the key variables “team identification”, “club reputation”, “CSR perception” and “perceived dialogue strategies” which will be explained in more detail in the following section “Measured variables”. Additionally, this part included a short description of the concept “Corporate Social Responsibility” as it could not be assumed that every participant was familiar with the framework. In the final part the participants’ demographics were gathered.
Sampling Method and Composition of the Sample
The survey data was collected with a non-probability sample. This sampling method does not rely on random selection, meaning that the probability to be selected cannot be calculated
21 and is therefore not equal. A disadvantage of this method is the weak sample-to-population representativeness. Regarding the limited time frame and available resources, this method was most suitable. Furthermore, it was still likely to result in an adequate sample regarding size and decent levels of football knowledge because the participants were targeted based on being a football fan of either a Bundesliga or a Premier League club. Non-football fans were not considered in the study.
The sample was targeted via the exponential snowball system and the online survey was therefore distributed through different platforms. On the one hand it was published on various football-related groups on social media platforms like Facebook, LinkedIn and Reddit. On the other hand, it was shared among the researcher’s personal social system via email, Instagram and the messenger platform WhatsApp in England and Germany. Additionally, 19 fan
initiatives from Germany were contacted via email to achieve more variance in clubs amongst the supporters of the Bundesliga. Also, Premier League fan forums were contacted but the requests were rejected.
The final sample comprised a total of 373 respondents, however only 279 of the questionnaires were valid for further investigations. In total, 25% (n=94) of all cases were incomplete and thus had to be excluded. Overall, the participants were on average 34.11 years old (SD=12.43). The composition by gender shows a significant majority of male subjects - with 79.9% (n=223) they represent more than two thirds of all participants. In contrast, only 55 women took part in the survey (19.7%). Although this does not fully reflect the gender distribution of the population of European football fans, it can be considered as representative (Statista Research Department, 2016). Asked about their football fandom on Likert-scales ranging from 1-5, the participants see themselves as rather strong fans (M=4.13, SD=.784). In addition, more than half of the respondents (n=145) indicated that they were members of their favorite club.
22 Regarding the distribution of fans across the two investigated leagues, the dataset shows a clear imbalance in the sample. While 209 of the interviewees stated that they were fans of the Bundesliga, only 70 fans of the Premier League were surveyed. This can be attributed to the sampling method and the social environment of the researcher. However, regression analyses, which are conducted during the hypotheses testing, are robust against unevenly distributed groups. Among the supporters of the Bundesliga, fans of FC Bayern München (n=119), 1. FC Union Berlin (n=31) and Borussia Dortmund (n=18) were most frequently interviewed. The fans of FC Bayern München represent more than half of the Bundesliga sample, however the club has by far the most fans within the league. This sample inequality is also countered by the fact that the second most frequent fan group is from 1. FC Union Berlin, which is one of the smallest German professional clubs. Within the Premier League clubs most of the participants were fans of Liverpool FC (n=22), Arsenal FC (n=20) and Manchester United (n=15). Those three clubs are also amongst the most supported clubs of the Premier League.4
Observed Main Variables
In the following, the main variables observed for the research interest are going to be described in more detail.5
CSR Perception. The fan perception of football clubs’ CSR activities was measured as independent variable influencing a club’s reputation. The concept was based on scales of previous research by Bianchi et al. (2019) who investigated the impact of perceived CSR on corporate reputation. Their measured reliability had a Cronbach’s α greater than 0.8 for all items. Some items were derived from a scale from the authors Liu et al. (2019) who especially researched perceived CSR performance and its impact on fan-based patronage intensions in Chinese professional football (Cronbach’s α ranged from 0.84 to 0.90). The scales were then
4 For further information for numbers of fans interviewed of all clubs see Appendix C.
23 aligned with the researched CSR themes (Kolyperas, 2012; Kolyperas et al., 2015) of the conducted content analysis and showed a high agreement. The participants were asked to rate the CSR actions of their favorite football club with 9 items on a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree. The items were for example “I feel that the club is committed to ecological/sustainable issues” or “I feel that the club improves community sports development”.
Club Reputation. The club reputation was measured as dependent variable influenced by fans’ CSR perception. A corporate reputation scale was originally developed by Caruana (1997) and validated by Caruana and Chircop (2000) in a following study with adequate reliability scores (Cronbach’s α = 0. 85). Walker and Kent (2009) modified the existing scale to capture the parameters of team-level corporate reputation. Therefore, a sporting context adopted scale is used in this study. It includes items like “The football club I support has a strong management” or “The football club I support makes the right financial decisions”, which were answered on a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree.
Team Identification. Team identification (Team ID) was measured as moderating variable for the relationship between perceived CSR and corporate reputation. The adopted scale was based on a six-items scale by Wann and Branscombe (1993) which is widely used in sporting contexts (Jung, 2012). The six items were based on a five-point Likert scale with different characteristics depending on the question. The participants were for example asked “How important is being a fan of this team to you?”, which could be answered with options from 1=not at all important to 5=extremely important.
Perceived Dialogue Possibilities. This variable was handled as an independent variable influencing corporate reputation and perceived CSR and was researched exploratively. As no validated scale was found for the concept during literature research it had to be based on available item suggestions and was linked to investigated dialogue possibilities within the
24 content analysis of this study. A principal component analysis had to be carried out during the analysis in order to test the scale on validity and reliability and is reported in the subsequent results section. The items are based on suggestions of Monfort et al. (2019), who in turn based their considerations on involvement strategies respectively dialogical strategies by Morsing and Schultz (2006) as well as Colleoni (2013). The survey participants were asked whether their supported football clubs offer good possibilities for engagement regarding the clubs’ CSR strategies on different platforms (e.g. social media, participative website, official fan forum). Each item was rated on a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree.
Results
In the following section, the results for both studies, the content analysis and the online survey will be described. Both parts will first present the mere descriptive results of the two methods, subsequently the outcome for the three sub-research questions as well as the five hypotheses will be discussed.
Study 1: Football Clubs’ Communication about CSR
In general, it was found that all 24 investigated football clubs engage in various CSR activities. Compared to previous studies (e.g. Hovemann et al., 2011; Kolyperas, 2012), it seems that these activities have increased significantly in recent years. Football clubs appear to pay even more attention to CSR than they used to do ten years ago. However, the clubs differ in terms of their activities and their CSR communication channels. Additionally, they offer fans various opportunities to enter dialogue with the club regarding CSR measures but also vary slightly in those possibilities.
Regarding the clubs’ usage of basic publication platforms in both leagues in terms of CSR activities, it only seems to differ significantly in two aspects. While all football clubs publish their CSR activities through their official club websites, there are considerable differences in
25 the way they communicate their actions in official reports. In the German Bundesliga, 42% (n=5) of all clubs publish their CSR results in a report, whereas only one Premier League club reports on CSR according to international standards. On the other side, English clubs present their CSR much more often in annual reports (Premier League: n=8 / Bundesliga: n=1) and publish additional ones like “Gender Pay Gap” reports, while no German team did so. This could be due to the financial corporate structure of many English clubs, which often forces them to publish annual reports.
When it comes to the execution of their CSR efforts, the clubs of the two leagues differ only slightly from each other. Almost all clubs gave the impression that many activities are managed by their own internal CSR department (n=23). In addition, over 90% of the English clubs have their own foundation, through which many other specific projects are realized. Nevertheless, also 8 of the 12 German clubs have their own charity foundation.
CSR Themes
The first sub-research question aimed at investigating what kind of CSR themes football clubs from both leagues include in their activities. For this purpose, results of previous studies had to be refreshed in order to be able to draw a better comparison with the results of the subsequent survey. The outcomes show that football clubs address various topics and differ from businesses of other economic sectors.
When analyzing the data, it can be seen that the football clubs basically address the same issues as they did several years ago. However, the number of clubs involved in specific CSR issues has increased (Figure 2). Especially the themes Community Football,
Charity/Fundraising, Social Inclusion/Diversity/Equity as well as Health Support were addressed by all 24 investigated clubs. Additionally, all English clubs engage in the topics Education through Sport and Fan-led Initiatives while slightly less German clubs did (83%).
26 Furthermore, the content analysis identified three additional topics that are frequently addressed by the clubs: the Covid-19 Pandemic, Safeguarding and Historical Reappraisal. The topic Covid-19 Pandemic was taken particularly seriously by all 24 clubs. Every football club had either its own section on its website, information pages or numerous support
initiatives to help fighting the pandemic. This can be mainly explained by the current social relevance of the issue and the devastating consequences of the pandemic, not only for
economic and health sectors worldwide, but also for European football. Safeguarding was an important part of the CSR topics especially for English clubs (83%). They significantly addressed the topic more frequently than their German counterparts (25%). The issue often involved protection of children or actions against modern slavery. In contrast to this, 33% of the German football clubs were concerned with historical topics in terms of reappraisal and educational work regarding the period of National Socialism in Germany. Interestingly, also two English football clubs put the topic on their agenda and communicated actively about it.
Two issues that are among the main CSR themes in other economic areas are addressed rarely by some of the football clubs. While 92% of the investigated German clubs deal with the topic of environmental protection - at least partially -, only slightly more than half of the investigated Premier League clubs communicate about this topic (58%). Also, internal CSR, to which many companies nowadays assign a very high priority on their CSR agenda, is often not explicitly communicated by football clubs. While 8 of the English clubs address the issue and publish for example "Gender Pay Gap” reports, only 6 of the German clubs communicate on the topic, which corresponds to only 50%.
27
CSR Communication Channels and Dialogue Strategies
The second sub-research question aimed at investigating the communication channels used by the football clubs to communicate their CSR activities. It was also examined which
opportunities fans were given for entering dialogue with the club regarding CSR issues. In total, English clubs used a greater variance of communication channels linked to their websites to inform about their CSR initiatives. Especially their social media channels are utilized to communicate their activities. All Premier League clubs use YouTube, Twitter and Facebook as communication channels, however, the popular platform Instagram is only used by 67%. The Bundesliga clubs, on the other side, communicate most frequently through Facebook and Twitter (n=12), but publish slightly less often about CSR on Facebook (83%) and Instagram (58%). Besides the previous mentioned fact that all clubs communicate through their official websites, all English clubs published press releases on some of their CSR
engagements (Bundesliga: n=10). The greatest difference can be seen in the use of the official
0% 25% 33% 50% 25% 83% 92% 83% 83% 100% 100% 100% 100% 100% 17% 0% 17% 67% 83% 83% 58% 100% 100% 100% 100% 100% 100% 100% 0% 20% 40% 60% 80% 100% Other Arts & culture Historical reappraisal
Employees Safeguarding Promotion of sport Environment & sustainability
Fan-led initiatives Education through sport
Covid 19 Pandemic Health, support & physical activities
Social inclusion, diversity & equity Charity / Fundraising Activities
Community Football
CSR Themes
Premier League (n=12) Bundesliga (n=12)
28 Club TV. While all Premier League clubs published videos about their CSR activities via this channel, only 7 of the Bundesliga clubs were doing so.
The second part of the sub-research question aimed at investigating possible dialogue strategies applied by the football clubs to engage with their supporters. Unsurprisingly, many football clubs are giving fans the opportunity to get in touch through their social media channels as this is a popular form of stakeholder involvement nowadays. In total, 18 of the 24 analyzed websites contained links to social media articles on CSR activities, with the English teams slightly ahead (n=10). In contrast, German clubs more frequently gave the opportunity to share CSR communication via social media (n=12 / Premier League: n=10). On Twitter, for instance, fans then have the opportunity to tag their club in the post and thus communicate their opinion to the club. Furthermore, 4 German and 6 English clubs had official debating forums, in which fans can discuss various topics not only with other supporters but also with the club. Additionally, 6 German clubs reported on ‘round tables’ with fans about CSR activities to discuss improvements or changes. Only 3 Premier League clubs were coded applying this direct communication strategy. Interestingly, the content analysis revealed various other methods, applied by the clubs, to get feedback from the fans. For instance, a German club offered its supporters a digital chatbot to get in contact with the club, also on CSR engagement, while another club interviewed the fans about their opinion on CSR topics. Three English clubs offered their fans the possibility to contact them via email especially for questions regarding their CSR engagement.
29
Figure 3: Dialogue Strategies applied by the football clubs (N=24).
Study 2: Fan Perceptions of Club CSR Data Characteristics
Before the analysis of the online survey could be performed, the data had to be cleaned and three of the measured main constructs had to be checked for validity and reliability as their items were adapted by the researcher. In order to extract the most independent factors, a Principal Component Analysis (PCA) was conducted for the constructs Team Identification, Perceived CSR and Perceived Dialogue Strategies. The Kaiser-Meyer-Olkin (KMO)
measures of sampling adequacy were .84 (Team ID – 6 items), .86 (CSR – 9 items) and .83 (Dialogue – 5 items) representing a good factor analysis for every construct. The Bartlett’s test of Sphericity was significant (p < .001) for all three analyses. Both, KMO and Bartlett’s test showed that the items were suitable for performing a PCA on every construct. The analyses indicated that all items correlated positively with the first component, which was also supported by each scree plot. Only factors with eigenvalues ≥ 1 were considered. The factor of Team ID (highest loading .68) explained 54% of variance with a good reliability
8% 17% 50% 17% 33% 67% 100% 0% 17% 25% 58% 50% 83% 83% 0% 20% 40% 60% 80% 100% Comment sections for articles, press releases, etc.
Participative websites Reporting or press release on meetings with fans (round
tables) on CSR topics
Other Links to debating forums Links to Social Media posts about CSR activities Buttons for sharing articles via Social Media
Dialogue Strategies
30 with Cronbach’s alpha = .82. The factor of Perceived CSR (highest loading .62) explained 46% of variance with a Cronbach’s alpha = .84 and has therefore a strong reliability. For the last construct, Perceived Dialogue Strategies, also only one factor was extracted (highest loading .72) with explaining 63% of variance and a Cronbach’s alpha = .85, indicating a strong reliability as well. After performing the PCAs, the items of the constructs were combined into the three variables as means.
Descriptives of the Variables
A general summary of the descriptives and the test for multicollinearity for all main and control variables is provided in Table 1. The descriptives were analyzed for both leagues in order to be able to determine a first tendency of the effects. Overall, the analysis indicated no obvious differences between Premier League (PL) and Bundesliga (BL) fans. The aggregated CSR perception score had a mean slightly above the midpoint for both leagues (PL: M = 3.85, SD = .56 / BL: M = 3.99, SD = .59). The dependent variable Club Reputation scored the highest on a 5-point Likert scale (M = 4.17, SD = .60 / M = 4.22, SD = .78), while the
moderator, Team ID, was rated as weakest just above the midpoint (M = 3.45, SD = .91 / M = 3.43, SD = .87). The new measured construct Perception of Dialogue Strategies was biased towards the middle category for both leagues (M = 3.11, SD = .79 / M = 3.19, SD = .77). The control variable Football Fandom, measuring to which degree a supporter considers himself as a fan, scored also considerably high indicating that the participants see themselves as rather strong fans (M = 4.16, SD = .78 / M = 4.11, SD = .78). When looking on the correlation matrix of the independent variables, all correlations were found to be well below 0.7, indicating that no multicollinearity was present (e.g. Kaplan, 1994) (Table 1). Additionally, for confirming the results of the correlation matrix, the variance inflation factor (VIF) was checked, showing also no signs of multicollinearity. The VIF scores were well below the commonly used problematic value of 10 and the R² varied in the acceptable range (.46 to .69)
31 (e.g. Hair et al., 1995). However, when looking at the other assumptions for
moderated/mediated regression analyses not all of them were met. When examining the scree plots of the standardized residuals for the multi-regression model, the outcome indicated that no heteroskedasticity was given. This was countered by using the robust standard error HC4 within the analyses (e.g. Hayes & Cai, 2007).
Table 1: Descriptives and correlation matrix for main and control variables per league (N = 279). Descriptives Correlation Matrix Construct Mean Standard
Deviation 1 2 3 4 5 1. Perceived CSR 3.85 / 3.99 .56 / .59 1.00 2. Team ID 3.45 / 3.43 .91 / .87 .339** 1.00 3. Dialogue Strategies 3.11 / 3.19 .79 / .77 .522** .204** 1.00 4. Football Fandom 4.16 / 4.11 .78 / .78 .252** .637** .058 1.00 5. Club Membership - - -.112* -.583** -.090 -.420** 1.00 DV: Club Reputation 4.17 / 4.22 .60 / .78 - - - - -
Notes: DV: Dependent variable; Descriptives: Premier League / Bundesliga; *p<.05; **p<.001.
Hypotheses Testing
To answer the five hypotheses and further sub-assumptions, multi-regression analyses and moderator and mediator analyses were conducted.
The first hypothesis investigated the positive relationship between the predictor Perceived CSR and the dependent variable Club Reputation. This was tested in a multiple linear
regression analysis (MRA) together with the two control variables (Football Fandom, Club Membership) and the third hypothesis which proposed that also the degree of fans’ perception of dialogue possibilities regarding club CSR has a positive impact on club reputation (Table 2).
The model was overall significant (F(4, 274)= 23.194, p < .001) and can therefore be used to predict club reputation based on the four independent variables. The R² for the overall model was .25 (adjusted R² = .24), explaining 25% of variance in club reputation and indicating a high goodness-of-fit. The results demonstrate that only Perceived CSR,
32 b* = 0.46, t = 5.82, p < .001, 95% CI [0.35, 0.71], has a significantly strong and positive association with Club Reputation. The other variables Dialogue Strategies, b* = 0.03, t = 0.62, p = .54, 95% CI 0.07, 0.14], Football Fandom, b* = 0.06, t = 0.80, p = .42, 95% CI [-0.08, 0.18] and Club Membership, b = 0.02, t = 0.27, p = .79, 95% CI [-0.13, 0.17] had no influence on the dependent variable. For every unit increase on the Perceived CSR scale, the reputation of a football club increases by 0.463 units. These results lead to a confirmation of H1, whereas H3 must be rejected as no significant impact of dialogue strategies on club reputation could be observed.
Table 2: Multiple regression model to predict Club Reputation (N = 279).
Unstandardized Coefficients Standardized Coefficients
Independent Variables B Robust Std. Errora Beta ta Sig.a
Constant 1.742 .410 - 4.244 .000*
Perceived CSR .330 .092 .463 5.819 .000* Perceived Dialogue Strategies .034 .056 .039 .616 .539
Football Fandom .053 .066 .060 .801 .424 Membership .021 .076 .015 .272 .786 R² .253 adjusted R² .242 F(df=4, 274) 23.194 .000* Notes: a. HC4 method.; *p<.001.
In a second step, a moderated simple regression analysis was calculated on H1 with the Process plug-in by Hayes (2017) to examine whether the effect is moderated by the supported league and is stronger for the Bundesliga than for the Premier League (H1a) (Table 3). The moderator was coded as a dummy variable with 0 = Premier League and 1 = Bundesliga. Contrary to the assumption of H1a, the results reveal that there is no statistically significant interaction effect between the supported league and perceived CSR on club reputation, b* = 0.14, t = 0.73, p = .464. This means that the relationship is not stronger for Bundesliga fans compared to supporters of the Premier League and therefore Hypothesis 1a must be rejected.
Hypothesis 2 assumed that the degree of a fan’s identification with his supported team moderates the effect of CSR perception on club reputation in such a way that the effect is
33 weaker for fans with higher team identification. It was also expected that this moderation effect is stronger for Bundesliga than for Premier League fans (H2a) (Table 3). The overall analysis revealed that Team Identification has no statistically significant interaction effect on the relationship of perceived CSR and club reputation, b* = 0.19, t = 0.99, p = .323.
Therefore, the influence does not differ significantly between fans with high or low team identification and the results even show that the effect between CSR and club reputation is slightly stronger for fans with high team identification. The second hypothesis (H2) is thus not approved. When looking at the results for each league, they also demonstrate a positive but not significant moderation for both leagues, with being slightly stronger for fans of the
Bundesliga, b* = 0.20, t = 0.91, p = .363, than for the Premier League, b* = 0.18, t = 0.13, p = .894 (Table 3). Subsequently, Hypothesis H2a must also be rejected.
Table 3: Moderated simple linear regression models for H1a, H2 and H2a. Interaction term Beta
Coefficient Robust Std. Errora t Sig. Rb R²,b Change in R²,c H1a Per. CSR x League Dummy (N=279) .139 .189 .734 .463 .502 .252 .003 H2 Per. CSR x Team ID Dummy (N=279) .185 .187 .990 .323 .506 .256 .006 H2a Per. CSR x Team
ID Dummy (n=70/n=209)
(.180/.201) (1.35/.220) (.133/.911) (.894/.363) (.450/.523) (.203/.273) (.005/.007)
Note: Dependent variable = Club Reputation; H2a: (PL/BL); a. HC4 method, b. overall model, c. change through Interaction; p = .05.
Hypothesis 4 expected a direct positive relationship between perceived dialogue possibilities and perceived CSR. This was analyzed with an MRA including two control variables (Football Fandom, Club Membership). The results of the MRA reveal that perceived dialogue possibilities positively predict perceived CSR (Table 4).
The overall model was statistically significant (F(3, 275)=50.906, p < .001) with an R² .36 (adjusted R² = .35) and therefore explains 36% of variance in perceived CSR. The results demonstrate that the construct Perceived Dialogue Possibilities, b* = 0.54, t = 11.06,
34 p < .001, 95% CI [0.33, 0.48], has a significant, strong and positive association with
Perceived CSR. The control variable Football Fandom was moderately significant with a positive effect on CSR, b* = 0.23, t = 4.42, p < .001, 95% CI [0.10, 0.26], while Membership had no significant influence on CSR, b* = 0.01, t = 0.22, p < .001, 95% CI [-0.11, 0.14]. Hypothesis 4 can therefore be confirmed.
Table 4: Multiple regression model to predict Perceived CSR (N = 279).
Unstandardized Coefficients Standardized Coefficients Independent
Variables
B Robust Std. Errora Beta
ta Sig.a
Constant 1.921 .236 - 7.809 .000*
Perceived Dialogue Strategies .406 .040 .536 11.057 .000* Football Fandom .176 .042 .234 4.422 .000* Club Membership .014 .062 .012 .224 .823 R² .357 adjusted R² .350 F(df=3, 275) 50.906 .000* Notes: a. HC4 method.; *p<.001.
The last hypothesis (H5) assumed a mediation effect of CSR perception between perceived dialogue strategies and corporate reputation based on H3 and H4. The effect was tested with a simple mediation analysis with PROCESS by Hayes (2017). The results demonstrate that the relationship between the perception of dialogue strategies and the reputation of a club is fully mediated by the perception of CSR, indirect effect b = 0.23, 95% CI [0.16, 0.32] (Table 5). Perceived Dialogue Strategies significantly predicts Perceived CSR, b* = 0.55, t = 10.63, p < .001, 95% CI [0.34, 0.50] and Perceived CSR significantly impacts Club Reputation, b* = 0.481, t = 6.64, p < .001, 95% CI [0.39, 0.72], but the direct effect between dialogue strategies and reputation is – as also demonstrated with H3 – not significant, b* = 0.03, t = 0.531, p = .596, 95% CI [-0.08, 0.14]. Although some researchers see a problem for a mediation in the insignificance of a direct effect, others argue that significance must not be given to proof a
35 mediation via an indirect effect (e.g. MacKinnon, 2008). In conclusion, the results of the mediation analysis lead to a confirmation of assumption H5.
Table 5: Simple mediation analysis (N=279). Regression term Beta
Coefficient Robust Std. Errora t Sig. R R² F(df)a Direct effect IV on DV .029 .054 .531 .596 - - - IV on mediator .552a .039 10.626 .000** .552 .305 112.92** (1, 277) Mediator on DV .481 .084 6.636 .000** .500 .250 27,318** (2, 276) BootLLCI BootULCI Indirect effect IV on DV .232 .155 .320 - - - -
Notes: DV: Club Reputation, IV: Perceived Dialogue Strategies, Mediator: Perceived CSR; a. HC4 method; *p < .05, ** p < .001.
Figure 4: Conceptual model with effect sizes. Note: *significant at p < .001 level; ns: non-significant.
Alignment of CSR Communication and Stakeholder Perceptions
The last sub-research questions aimed at investigating the alignment between disclosed CSR engagement and communication by the football clubs and their perceptions by the fans. The alignment was tested by comparing the percentage of clubs engaging in the CSR themes and using the applied dialogue strategies with the mean score on the perception scales of each
36 single CSR and dialogue strategy item. In doing so, it was investigated whereas there is a congruence or gap between perception and actual commitment and whether fans and clubs have the same rank order for the items.
The results of the alignment testing for CSR engagement and perception show high similarities for some CSR themes (Table 6). While community football, charity/fundraising and social inclusion/diversity/equity were addressed by all investigated clubs (N=24), they also were perceived the strongest by the fans with a mean score ranging between 4.06 to 4.26 for Premier League supporters and 4.19 to 4.25 for Bundesliga fans. Interestingly, although all clubs engage intensively in health and physical activities, they are not that strongly perceived by the fans with only being on 4th place for Premier League fans and even at the penultimate place amongst the Bundesliga fans. There is also a large gap between disclosed sustainability and environmental engagement of the Bundesliga clubs and the fans’ perception of it.
Although 11 of the 12 clubs had initiatives regarding such issues, they were perceived the weakest by the Bundesliga fans amongst all CSR themes (M=3.43). In contrast, the clubs’ internal CSR regarding their employees is perceived quite strong, with ranging at the 5th place among Premier League fans (M=4.00) respectively 4th place for Bundesliga supporters (M=3.98), although in total the clubs’ actual engagement was the lowest out of all CSR themes (58%).
37
Table 6: Alignment of disclosed CSR engagement and fan perception.
Overall (N=279) Premier League (N=70) Bundesliga (N=209) CSR Themes
(sorted by importance among the clubs)
Percentage of football clubs addressing the topic Perception by the fans - Mean Score (Rank order) % M % M 1. Community Football 100% 4.21 (2) 100% 4.26 (1) 100% 4.19 (3) 2. Charity / Fundraising 100% 4.24 (1) 100% 4.21 (2) 100% 4.25 (1) 3. Social inclusion,
diversity & equity
100% 4.19 (3) 100% 4.07 (3) 100% 4.23 (2)
4. Health, support & physical activities 100% 3.89 (5) 100% 4.06 (4) 100% 3.83 (7) 5. Fan-led initiatives 92% 3.84 (6) 100% 3.67 (6) 83% 3.90 (6) 6. Environment and sustainability 75% 3.42 (8) 58% 3.39 (7) 92% 3.43 (8) 7. Promotion of sports (non-football) 83% 3.73 (7) 83% 3.04 (8) 83% 3.96 (5) 8. Employees 58% 3.99 (4) 67% 4.00 (5) 50% 3.98 (4)
When looking at the alignment of the football clubs’ usage of dialogue strategies and the fans’ perception of it, the results indicate differences in actual applied strategies and its perception (Table 7). Social media is most often used by the clubs for giving their
stakeholders the opportunity to interact with the club regarding its CSR initiatives (75% total). Accordingly, fans perceive interaction possibilities through Social Media high, with ranging at the first (Premier League: M=3.29) respectively second place (Bundesliga: M=3.32). Although the clubs provide debating forums second most often for interaction regarding CSR (PL: 50% / BL: 33%), this dialogue possibility is not perceived as strong by the fans and therefore ranges on the last place for Premier League fans (M=2.97) and at the third for Bundesliga supporters (M=3.10). Surprisingly, although no Premier League club offered the possibility for CSR fan engagement on its website (e.g. comment sections), fans’ perception ranked on the third place amongst Premier League fans (M=3.06).
