Introduction

Research indicates that sports participation is socially stratified, with people from a lower socio-economic background being less likely to participate in sports (Hartmann-Tews 2006; Vandermeerschen et al. 2016; Van Bottenburg et al. 2005; Van Tuyckom and Scheerder 2010). Despite 40 years of Sport-for-All policies—at least in many European countries—sports participation is still positively associated with socio-demographic factors such as education or income. For example, results from the Eurobarometer 2014 indicate that across all member states of the European Union, 68 % of the adult population (15 years and older) who quit studying at the age of 15 or earlier never participates in sports, whereas this is 27 % among those who continued studying until the age of 20 or longer (European Commission 2014). Moreover, specific societal groups, such as people with a disability, people from a different ethnic-cultural background and people living in poverty, are in a disadvantaged position with regard to active participation in sports, experiencing various barriers to sports (Collins 2004; O’Driscoll et al. 2014) and to membership of nonprofit sports clubs (Wicker and Breuer 2013).

Though claims sometimes lack sufficient empirical evidence (see e.g. Coalter 2007a), sports participation tends to be associated with numerous benefits, f.i. in terms of health or social capital (Hoye et al. 2010; Nicholson and Hoye 2008). In addition, sports can be considered as a social right, which is reflected in several international policy documents, such as the European Sport for All Charter (COE 1975; COE, CDS 1980), and the United Nations’ International Charter of Physical Education and Sport (UN, UNESCO 1978). As a consequence, the unequal (opportunities for) participation pose(s) problems both in terms of policy efficiency as well as social justice. Yet, it seems unlikely that the participation gap will disappear by itself (Taylor 2001). If we are to reduce this gap, structural, collective efforts are needed. As nonprofit sports clubs are increasingly expected to fulfil wider objectives, both in terms of sports policy (Enjolras 2002; Feiler et al. 2015; Harris et al. 2009; Vos et al. 2012) as well as social policy (Persson 2008), a question to be asked is to what extent sports clubs can induce change in this regard, more particularly in reducing the participation gap in organised sports.

As indicated by Skille (2009), there is a lack of research at the club level, with the sports clubs’ representatives playing an important, active role in the realisation of sports policy. The aim of this paper is two-fold. First, we want to investigate both how sports clubs perceive their role and responsibility towards socially disadvantaged groups and how they act on it. In other words, the first aim is to get insight in the current situation regarding the role of sports clubs for reaching social inclusion in sports. Second, we will explore what predicts the position of the sports clubs in this regard. For this second question, we focus on their initiatives towards one particular socially disadvantaged group, namely people living in poverty.

The focus on people in poverty in the second part of this study is motivated by recent policy developments in Flanders (Belgium), the research context of this paper, with the inclusion of this group increasingly reaching the (sports) policy agenda, and at the same time still posing many challenges to both policy makers and sports providers (Collins 2004). Understanding the factors that contribute to the presence or absence of club efforts directed towards this group can be helpful for correctly estimating, and possibly strengthening the role that can be played by the sector of club-organised sports in reducing the participation gap for people in poverty. Poverty is a widespread problem, as 17 % of Europeans have an income below their country’s poverty line (European Commission 2014). In the literature as well as in policy, awareness has grown that poverty is to be understood in relative and relational terms. In other words, as expressed in the seminal work of Townsend (1979), poverty is not merely about a lack of resources, but about a lack of resources relative to standards in society and hampering participation in customary activities, amongst other things.

The Changing Role of the Voluntary Sector: Towards Wider Goals for Sports Clubs?

Over the past decades, the field of sports has significantly gained importance in society, an evolution which is generally referred to as the ‘sportification of society’ (see Crum 1991, p. 15). The more central position that sports has come to take, is also reflected in sports policy. Whereas the field of sports was traditionally dominated by the voluntary sector, a ‘governmentalisation’ has been identified in the literature (see f.i. Bergsgard et al. 2007), indicating governments’ increasing involvement in terms of regulation, financial investments, promotion, etc. (see f.i. Green 2005; Green and Collins 2008; Houlihan 2005). The so-called instrumentalisation of sports (see f.i. Grix 2010; Vos et al. 2011), i.e. the increased use of sports as an instrument to attain broader policy goals, is also characteristic of this larger process of governmentalisation. The changed approach towards sports on behalf of governments affects the position of the third sector, as expectations of governments towards the voluntary sector, and more specifically towards nonprofit sports clubs, have grown. Sports clubs are expected to play a role in promoting sports participation, and/or to contribute in achieving additional, nonsports-related policy goals (see f.i. Harris et al. 2009; Persson 2008; Van Bottenburg 2011; Waardenburg 2016).

As argued by Adams (2011), Stenling and Fahlén (2014), and Waardenburg (2016), amongst others, governments increasingly assume that sports clubs are able and willing to act as policy implementers, aiming to accomplish wider, external goals. Social inclusion, which is the focus of this paper, is one of these goals. Other examples are health promotion (Meganck et al. 2015), enhancing social cohesion and reducing anti-social behaviour (Bloyce and Smith 2010; Coalter 2007a; Hoye et al. 2010). However, reality is not as straightforward (Skille 2008, 2009), and both the willingness and the capacity of nonprofit sports clubs to deliver policy goals is questioned in the literature (see f.i. Balduck et al. 2015; Harris et al. 2009; Vos et al. 2011; Wicker and Breuer 2013), due to the nature of sports clubs as organisations. Institutional theory, a theoretical framework which has been referred to in many studies in the literature on sports management and sports policy, and particularly focusing on the relation between state and voluntary sports organisations (see f.i. Slack and Hinings 1994; but also Skille 2009; Waardenburg 2016, amongst others), brings insight in this regard, as will become apparent in the next section. Yet, other important insights from this field of research will be mentioned as well.

Sports Clubs as Part of an Organisational Field: An Institutional Perspective

According to institutional theory (DiMaggio and Powell 1983; as well as Edwards et al. 2009; Meyer and Rowan 1977; Scott 1987; Zucker 1987, amongst others), organisations are characterised by unclear goals, and therefore rely on, and adapt to their environment—or their organisational field—in determining their course of action. Put differently, organisations are influenced by their institutional context. Institutions should be understood here as values, ideas and general ways of doing things, which determine what could or should be done, and how it should be done (Meyer and Rowan 1977; Oliver 1992). Following this theoretical perspective, organisations (i.c. sports clubs) will take over organisational behaviour from their environment to gain legitimacy, which causes organisations (clubs) to become more similar over time. Three types of pressures will lead organisations to become isomorphic, more particularly coercive pressures (resulting from power relations), mimetic pressures (copying successful behaviour of other organisations) and normative pressures (resulting from professionalism, taking over actions as taught through educational institutions, by specialists,…) (DiMaggio and Powell 1983). Translating this to our context of study, it implies that sports clubs are largely affected by their institutional context when determining their course of action. For example, clubs are likely to be influenced by what is ‘expected’ from them by national or local governments in terms of social policy. This influence can be exerted through rules, regulations and conditioned subsidies, but also through communication, advice and support towards clubs, or simply by examples set by the government. At first sight, pressures emanating from the government could be considered as a type of coercive isomorphism. However, normative and/or mimetic isomorphism might be a more accurate description (Skille 2009; Waardenburg 2016).

However, clubs are likely to be influenced by other sources as well, in addition to governmental pressures. The different values as transmitted by isomorphic processes come together in what is called the dominant logic, a concept put forward by Bettis and Pralahad (1995) and defined as a manifestation of shared values among actors constituting an organisational field (Fahlén and Karp 2010; Stenling and Fahlén 2009). Investigating sports clubs in Sweden, Stenling and Fahlén (2009) identified three dominant logics in the clubs: the sport-for-all logic, the result-oriented logic and the commercialisation/professionalisation logic. Yet, they observed that there is an order in these logics, with the sport-for-all logic being overshadowed by both other logics. They speak of the ‘hungry beast of result orientation and commercialization’. Also Fahlén and Karp (2010) conclude that ‘sports clubs are first and foremost concerned with competitive club sport and not so much with government initiatives for health or social integration’. Based on his research in Norway, Skille (2011) reached a similar conclusion, namely that the main sport convention in clubs is competitiveness. Competitiveness comes together with principles such as elitism, selection and exclusion, which are contradictory to social goods. Skille (2011) argues that the realisation of Sport for All and/or sports for social goods might be difficult to accomplish, since competitiveness is simultaneously pushing in the opposite direction.

Another element which may further hamper the process of implementation in terms of social goals, is the stability and inertia of organisations, as also mentioned in institutional theory. This seems to be the case in this context as well. For example, Fahlén and Karp (2010) observe that legitimacy seems first and foremost acquired by adhering to customary values, norms and traditions, rather than by abiding to new rules. This is however not to say that change is not possible. As emphasised by Stenling and Fahlén (2014), new sports clubs are being created, and also within existing clubs, change can occur, and new policy directions can be chosen. Along the same line, it is of fundamental importance to acknowledge that, even though there are isomorphic processes, there is still a heterogeneity among sports clubs (Coalter 2007b; Nichols and James 2008; Nichols et al. 2012; Stenling and Fahlén 2014), and expectations in terms of policy implementation should be adjusted accordingly (Stenling and Fahlén 2014).

According to Stenling (2013), a decisive factor for the potential of policies is the recognition that the response of sports clubs to new ideas is determined by their existing activities, or put differently, by the local context (Skille 2008). Policies should therefore not be in complete conflict with the self-identification of the clubs, or should be part of wider process of change which is already taking place in the institutional context. The question remains, then, whether this is the case here. Several authors have noted that sports clubs are first and foremost oriented towards their members and their members’ interests, and sports clubs’ goals are mainly introverted rather than being focused on the broader society (Ibsen and Seippel 2010; Nagel 2008; Reid 2012), which is also reflected in the motivation of volunteers (Allison 2001; Coalter 2007b; Harris et al. 2009). The demand to include wider social goals such a social inclusion might therefore be at odds with the current self-identity of the club. On the other hand, a research of Skille (2009) shows that sports club representatives do respond—or attempt to respond—to requirements in their local communities in that sense. However, the challenge emanating from the local community—in this case adapting to youth from the neighbourhood—was later perceived as too demanding by the club representatives.

Too Much Pressure?

According to some scholars, linking exceedingly high expectations with regard to the implementation of governmental goals may even be harmful, as it is said to carry the risk of putting too much pressure on nonprofit sports clubs (Harris et al. 2009; Morgan 2013). Too high demands might scare off volunteers when expectations are too different from their own motivations (Harris et al. 2009), while the recruitment of volunteers often already presents an obstacle for sports clubs (Seippel 2004). Human resources are the most pressing problem in the view of sports clubs, followed by financial matters (Seippel 2004).

Nevertheless, there are also arguments in favour of an enhanced involvement of sports clubs in the attainment of policy objectives such as social inclusion. Sports clubs are being supported with public resources, which would imply an additional social responsibility for the sports clubs (see f.i. Rulofs 2012). By the same token, it can be said that the integration of all citizens is a collective responsibility, of all members of society, and not merely of the government. We can also refer to the concept of corporate social responsibility (CSR) in this regard. Persson (2008) argues CSR should increasingly be used in the context of sports organisations. This idea could be extended to voluntary sports clubs as well. Skille (2009) argues that, while governmental expectations might not be sufficient to impose a policy (top-down), it might still trigger and encourage bottom-up initiatives (Skille 2009, p. 76–77).

The Organisational Culture of Sports Clubs

The implementation of wider social goals such as the social inclusion of disadvantaged people not only depends on the willingness and ability of the club to take inclusionary measures, but also on how the club is perceived by people in a disadvantaged position themselves. Referring to Doherty and Chelladurai (1999), Hanlon and Coleman (2006) argue that the success of the eventual recruitment and retention of disadvantaged groups (in their study, culturally diverse people) is dependent upon the organisational culture, and more precisely whether it is focused on diversity versus similarity. Doherty and Chelladurai (1999) have underlined the existence of these two different types of organisational culture (with a continuum in between), and claim that the culture of sports organisations has typically been one of similarity, which implies members are expected to share social values, cultural symbols and differences are reduced to a minimum. Hanlon and Coleman (2006) have found that the majority of sports clubs function based on organisational similarity, which can negatively affect the capacity to include minority groups. Rulofs (2012) supports this argument as well, stating that sports clubs will formally be open to everyone, whereas informally, subtle processes of exclusion are at work. It is the culture of the club, the developed routines and habits that determine whether or not new members will be able to feel a sense of belonging (Elling and Claringbould 2005; Rulofs 2012). Yet, for sports clubs and their representatives the first goal is to act in the interest of current members (Thiel and Meier 2004). When confronted with diversity, sports club managers prefer to homogenise the members’ interests (Rulofs 2012).

Research Questions and Framework

In this paper, it is our aim to investigate to what extent nonprofit sports clubs currently take a solidaristic stance towards socially disadvantaged groups. Second, taking club diversity into account, it will be empirically verified what factors can be identified as predictors, either as facilitators or as barriers.

In the context of northwest Europe, where the implementation of governmental sports policies at the grassroots level relies on voluntary sports clubs (Bergsgard et al. 2007), and a blurring of boundaries between sectors is apparent (Lucassen and van der Roest 2011), on the one hand, efforts towards disadvantaged groups may be expected to occur. In Flanders—a region corresponding to the Rhineland model (see Albert 1991, 1992), which is characterised by an active government, the search for consensus, a long-term perspective and social entrepreneurship—there is a strong relation between municipalities and sports clubs, and voluntary sports clubs receive considerable support from the government, directly (subsidies) and indirectly (f.i. the use of accommodation) (Vos et al. 2011). Therefore, clubs may invest in social inclusion, not only because they consider it in their own interest, but as a way of complying with (perceived) normative expectations, responding to institutional pressures. On the other hand, based on the current literature, we expect that pressures to aim at wider social goals such as social inclusion, may well be overshadowed by other, stronger influences (cf. supra), which may affect the clubs’ solidaristic stance towards disadvantaged groups.

For the second question, we focus on one specific disadvantaged group, more specifically people living in poverty. Four main (sets of) predictors will be included in the analysis. They are related to (i) the organisational capacity of the club (Hall et al. 2003), (ii) their resource dependence (Pfeffer and Salancik 1978), (iii) the social composition of the club, and (iv) the presence or absence of competition.

As the capacity of the sports club to successfully contribute to social inclusion is questioned in the literature, the organisational capacity of the clubs is a main focus in the selection of potential predictors in the analysis. Organisational capacity is to be understood here as ‘the ability of an organisation to draw on various assets and resources to achieve its mandate and objectives’ (Doherty et al. 2014, p. 125). A sports club might consider social inclusion as its mandate, but fail to do so because of a lack of organisational capacity. Hall et al. (2003) have provided a framework for analysing the organisational capacity of nonprofit and voluntary organisations, distinguishing three main dimensions: financial capacity, human resources capacity and structural capacity. Financial capacity refers to the ability to develop and deploy financial capital. Here, we can think of revenues, expenses, assets and liabilities of an organisation. Human resources capacity includes paid staff and volunteers, as well as their competencies, knowledge, attitudes, motivation and behaviour. Human resources capacity is considered the key element by Hall et al. (2003), as it leads to the development of the other capacities. Structural capacity regards ‘the ability to deploy the non-financial capital that remains when the people from an organisation have gone home’ (p. 5) and comprises relationships and networks, infrastructure and processes, and planning and development. The framework of Hall and colleagues fits well with grassroots membership associations (Doherty et al. 2014) and has been successfully applied to sports clubs in previous research (see f.i. Balduck et al. 2015; Doherty et al. 2014; Wicker and Breuer 2013). The framework of Hall forms a key foundation of the current study.

Yet, in addition to organisational capacity, the theory on resource dependence should also be mentioned (Pfeffer and Salancik 1978), as it is considered fundamental as well to explain the strategic behaviour adopted by nonprofit organisations (Akingbola 2013). Organisations usually depend on external resources. While gaining financial strength, it limits their autonomy, as their ability to act independently is reduced. In other words, their resource dependence may reinforce institutional pressures, more particularly to comply with the demands of the external resource provider(s), in this case the local government. Applying this framework to sports clubs, Vos et al. (2011) have found a positive effect between the dependency on governmental subsidies and the adoption of subsidy conditions with regard to offers for target groups, amongst other things. Therefore, here too, we include the dependence on subsidies of the local government as a predictor in the analysis.

Third, the social composition of the club is also taken into account, and more specifically the presence of underprivileged people. In a study on the organisational barriers to inclusion from the perspective of recreation professionals in the United States, Allison and Hibbler (2004) observed that one of the barriers to inclusion was the inability to recognise the diversity in the community served. If some of the club participants are underprivileged, and sports clubs’ representatives are aware of it, the likelihood of lowering barriers for people in poverty may be higher.

Fourth and last, based on the findings of Fahlén and Karp (2010), Skille (2011) and Stenling and Fahlén (2009) with regard to the contradiction between competition (as dominant logic or main convention in many clubs) and the realisation of sports for social goods—both pushing in opposite directions—, and the dominance of competition as the main logic, we also include the presence or absence of competition as a predictor in the analysis.

Policy Context

This research was conducted among sports clubs in Flanders, the northern Dutch-speaking part of Belgium. Belgium is a federal state. There are three main levels of decision making, more particularly (i) the federal level, (ii) the three communities (Flemish community, French-speaking community and German-speaking community), and (iii) the three regions (Flemish region, Walloon region, Brussels Capital region). Being considered as a ‘culture-related’ policy issue, sports policy is the responsibility of the communities. As there are consequently considerable differences in sports policy between the communities, we focus on club-organised sports in Flanders, instead of Belgium as a whole. Currently, attention to the participation of socially disadvantaged groups is clearly present in governmental policy. In 2007, a Sport-for-All decree was issued by the Flemish government, allowing local sports authorities to take on a larger role in sports policy by applying the principle of subsidiarity, while at the same time allowing the Flemish government to influence the local sports policy agenda. In this way, local sports authorities have been obliged to make explicit efforts to reach disadvantaged groups. The choice of ‘target group’, however, was left to the sports authorities, in order to be adaptable to local needs. Since 2016, local authorities have been granted more autonomy and local policy priorities are no longer imposed by the Flemish government. While poverty has always existed in Flanders, people in poverty can be considered as a rather ‘new’ target group, which has also motivated our choice to focus on this particular group in this study.

Contrary to the government, the formal role of the third sector in terms of encouraging participation of socially disadvantaged groups has been limited so far. In Flanders, there are around 23,900 nonprofit sports clubs. This implies the region has a dense network, of around 391 sports clubs for every 100,000 inhabitants (Vos et al. 2011). In general, most sports clubs in Flanders are small in size, and focus on one particular sport. They are predominantly nonprofit, and heavily rely on voluntary work. Many sports clubs are affiliated with an umbrella organisation. Since 2003, sports federations have been offered the possibility to apply for additional subsidies from the Flemish government by taking initiatives for specific social groups. The Flemish government selects a particular social group; the group targeted by this measure changes every few years. Currently, the additional subsidies are granted for initiatives with regard to youth. Previously, people of a different ethnic-cultural origin, people with a disability and elderly people have been put forward as target group by the Flemish government. For these groups, the response of the federations has been rather low, especially for people of a different ethnic-cultural origin (Scheerder et al. 2012). Until now, people living in poverty have not been selected as a target group in this context. This is somewhat remarkable, given recent socio-economic developments—with an increased polarisation in society—and the heightened policy attention for this group in other policy domains.

The Flemish government also directly encourages the sports federations and sports clubs to promote an ‘ethically responsible’ sports practice. Topics advocated in this context by the Flemish context are (i) the Rights of the Child in sport, (ii) inclusion, (iii) respect for diversity, (iv) fair play, (v) the physical and mental integrity of the individual and (vi) solidarity (Flemish Government 2012). Research among 193 youth sports clubs in Flanders has shown that fair play is the most popular topic to be implemented in sports clubs (almost 6 out of 10 youth sports clubs agreed with the statement that they undertake concrete actions to promote fair play during sports practice), followed by integrity (Seghers et al. 2012). Inclusion, which is the focus of our study, is found to be the least popular. Only one out of four youth sports clubs reported that they undertake concrete actions to lower their barriers to participation for specific target groups such as children from low income families. Respect for diversity (e.g. antiracism, fighting discrimination or tolerance for homosexuality) occupies an intermediate position.

Data and Method

Sample and Instrument

The data for this study stem from a survey called Flemish Sports club Panel (FSP), collected in 2012 at the University of Leuven (Belgium) (Scheerder et al. 2015). A first edition of the survey was held three years earlier, in 2009. The 2012 survey can be considered as a second wave. For the current study, only the data from 2012 are used. The selection of the sports clubs was made through a stratified random sample of municipalities based on their socio-economic profile. They were asked to either contact the sports clubs in their municipality by sending an information letter provided by the researchers, or to provide the researchers with the addresses of the sports clubs. The letter sent to the sports clubs contained the link to the questionnaire, which was held online. However, paper questionnaires were available if sports clubs preferred. It was necessary to select the sports clubs through municipalities because in Flanders, there were no administrative data available providing an overview of all sports clubs with their relative size and sports. Another option would have been to contact the sports clubs through their umbrella organisation (sports federation). However, contact by municipality was preferred in order not to exclude any sports clubs in advance, as not all sports clubs are affiliated to a sports federation.

In total, the sample consists of 580 sports clubs, 285 of which had also responded to the 2009 survey; 295 sports clubs were ‘new’ in the second wave (from 30 municipalities). Among sports clubs that participated in the 2009 survey, 48 % participated again. For new sports clubs contacted by the researchers, the response rate was 27 % (in eleven municipalities). Eighteen municipalities, however, preferred to contact the sports clubs directly. For this group, the response rate cannot be calculated, as we do not know the exact number of sports clubs that were contacted. One municipality refused collaboration. In 81 % of the cases, the survey was filled out by the president or secretary of the club. In a minority of cases (19 %), the survey was completed by other representatives, such as the treasurer, or by several people at a time.

The general aim of the survey is to gain insight in the characteristics, the functioning and the positioning of sports clubs in Flanders. The 2012 wave covers seven overarching items, i.e. infrastructure, organisation, membership, health, diversity, financial situation and cooperation with partners. Information used in the current study was based on six of the seven sections of the survey.

Method and Operationalisation

The first, exploratory research question—concerned with how sports clubs perceive their role with regard to broader social goals such as social inclusion, and how they act on it—is answered through descriptive analysis. Three different components are considered here: (i) the opinions of the club concerning its role in promoting diversity or inclusion, (ii) the structural anchoring of diversity in its policy and functioning, and (iii) the concrete efforts with regard to diversity and social inclusion that are being made (e.g. in terms of lowering financial or socio-cultural barriers). For the first component, respondents were asked to what extent they agreed with the following statement: ‘it is the task of a sports club to encourage diversity among the members and to try to accomplish that disadvantaged groups in society also take part in the club’, on a scale from 1 to 5 (ranging from totally disagree to totally agree). The first and second component relate to diversity and social inclusion ‘in general’, without singling out a particular group. The third component, on the other hand, provides a picture of the sports clubs’ efforts by target group. The first target group mentioned are people living in poverty, which is the ‘beneficiary’ of primary focus in this study. However, to allow for comparison, results for other groups (i.c. people with a mental, physical or sensory disability, people from a different ethnic-cultural background and elderly people) are also presented. The results across all target groups are mentioned as well. No definition of the different target groups was given; the interpretation was left to the respondents.

For the second research question we rely on multivariate logistic regression analysis, preceded by bivariate analyses. As mentioned earlier, the clubs were selected through municipalities. This could have induced dependence in the data, with sports clubs within the same municipality possibly being more alike. We have controlled for this nested structure of the data during the analyses. Yet, we found a ‘unique’ postal code for 57 percent of the sports clubs in our sample, indicating that in most cases, there was only one sports club per municipality in the sample. As also the intraclass correlation per municipality was very close to zero, in our final model, we used standard logistic regression rather than multilevel modelling. Analyses were conducted using Stata SE 12 (StataCorp 2011). No problems with multicollinearity were detected.

The dependent variable regards the efforts sports clubs put into lowering barriers for people in poverty. In the regression analysis, it is operationalised as a binary variable. The variable has value 0 when no efforts are mentioned, 1 when efforts are done to lower (i) financial barriers, (ii) socio-cultural barriers, and/or (iii) to facilitate integration in the club. More specifically, the value 1 refers to a ‘yes’ on at least one of these three items, a zero is obtained if for all three items ‘no’ was answered (for the exact question, see Table 5). This implies that the effort is self-reported; the analysis is based on the statement of the club representative, rather than on an external check of the actual efforts being made. Table 1 provides more insight in this variable, by showing the number of items that were indicated by the respondents (financial barriers, socio-cultural barriers and/or facilitating integration). For comparability reasons, cases with missing values on one or several of the independent variables are not counted here, as these were excluded from the regression analyses as well. Approximately half of the clubs (46.4 %) did not report any efforts, whereas a quarter of the clubs (26.3 %) responded affirmatively on all three items. When only one item is mentioned (15.8 %), this is most often the financial barrier (26 out of 33 cases). Similarly, two items usually correspond to the financial barrier, in combination with one other (in 18 out of 24 cases). In total, 53 percent of the clubs have value 1 on the dependent variable. Yet, also regression analyses on the separate items will be presented.

Table 1 Lowering barriers: number of indicated items (financial, social/cultural and/or facilitating integration) (N = 209)
Full size table

The independent variables in the analysis relate to (i) the organisational capacity of the club, (ii) their resource dependence, (iii) the presence of underprivileged people and (iv) the relative focus on competition within the club. Table 2 shows how the framework of Hall et al. (2003) has been adopted in this study, by providing an overview of the variables that relate to the organisational capacity of the club for each dimension.

Table 2 Overview of the variables related to organisational capacity
Full size table

Concerning human resources capacity, ‘volunteers’ are people who take an active role in and support the functioning of the club, but in another way than is the case for a board member or trainer. It concerns people performing tasks like refereeing, maintaining infrastructure or material, serving in the cafeteria of the club, or paramedical staff. In the framework of Hall et al. (2003), the dimension of infrastructure and process capacity refers to the ability to deploy or rely on infrastructure, as well aspects related to the internal structure and day-to-day operations. Here, we focus on the availability of sufficient accommodation. With regard to external relationships, we concentrate on the presence or absence of a partnership with local sports authorities, because of the central role of the latter in local diversity policies. Lastly, the presence or absence of a policy plan, i.e. a plan containing strategic and operational goals of the club, is informative for the planning and development capacity of the sports club.

Resource dependence is included in the analysis by the percentage of total annual revenues coming from subsidies from the local sports authority. The presence of underprivileged people is captured in a binary variable. We do not have information on the relative share of underprivileged people in the club. It should be noted that the variable refers to the presence of underprivileged people according to the club representative. Finally, the variable linked to competition in a sports club, makes a distinction between sports clubs providing recreational sports, competitive sports, or both. Table 3 provides an overview of the variables in the analysis and the distribution of cases.

Table 3 Operationalisation of the variables and distribution of cases (N = 217)
Full size table

Results

Opinion, Structural Anchoring and Concrete Realisations

With regard to the statement that ‘it is the task of a nonprofit sports club to encourage diversity among the members and to try to accomplish that disadvantaged groups in society also take part in the club’, opinions were rather diverged, but only a minority (16 %) really disagreed. Approximately one-third of the respondents (34 %) gave a neutral answer. Half of the respondents (51 %) agreed with this statement. The mean score was 3.4 (N = 414, SD = 1.0). In the survey, respondents were also asked to what extent they agreed with the statement that ‘everybody is welcome at our club, regardless of a person’s social, cultural or ethnic background’. Over nine out of ten respondents (94 %) agreed to this statement, only 1 % disagreed and 5 % ticked the neutral answer. The mean score here was 4.5 (N = 421, SD = 0.7) (results not presented in a table). The findings suggest sports clubs do not want to exclude in advance, but opinions differ on the extent to which it is their responsibility to actively pursue diversity and social inclusion in their club.

In the survey, respondents were also questioned with regard to the formal, structural integration into the club’s policy and functioning of certain issues, amongst which diversity/inclusion. While the term ‘diversity’ was used in the survey, its description is very close to ‘inclusion’, since diversity was explained in the survey as ‘attention for disadvantaged groups such as people from a different ethnic-cultural background, people with a disability, elderly people, or people living in poverty’. When asked if their club paid additional attention to diversity/inclusion in its policy and functioning, 23 % responded positively (Table 4). The structural anchoring of diversity policy in the club is rather limited. In about one out of ten sports clubs, a responsible is designated within the board. Other ways of structurally integrating diversity/inclusion, such as organising specific actions or having a separate budget are even less common.

Table 4 Structural anchoring of diversity policy in the club (N = 525), in percentages
Full size table

A partially different picture arises, however, when looking at realisations of the nonprofit sports clubs at a more concrete level (Table 5). When questioned about the presence or absence of concrete efforts and initiatives regarding the people in a socially disadvantaged position, 57 % of the sports clubs state to make efforts to promote their activities towards at least one of the socially disadvantaged groups under consideration (people in poverty, people with a mental, physical or sensory disability, people from a different ethnic-cultural background and elderly people).

Table 5 Concrete efforts (% yes) of sports clubs regarding diversity and social inclusion
Full size table

Nonprofit sports clubs were also questioned about potential efforts for lowering barriers to participation. More particularly, club representatives were asked about their efforts or initiatives to lower financial barriers and social and/or cultural barriers, and to facilitate integration in the club. Six out of ten sports clubs state to make efforts to lower some barriers for at least one of the target groups. In other words, though the structural anchoring of inclusion policy is rather limited, in practice more efforts are put forth. Finally, about one out of five (22 %) sports clubs also stated to have a specific sports offer, exclusively for a particular group.

Yet, there are rather large differences between the potential target groups. People in poverty are most frequently mentioned when it comes to lowering barriers to sports participation. About one-third of the sports clubs indicate to make efforts to lower social and/or cultural barriers for people living in poverty, or to facilitate or heighten their integration within their club. Over four out of ten sports clubs state to take initiatives to lower financial barriers for people in poverty. Promotion of club activities, on the other hand, is most often targeted on elderly people (44 %), though people in poverty are also frequently mentioned here (30 %). In reality, there is of course often some overlap between the different target groups.

Focusing on people in poverty, approximately half of the clubs (48.9 %) state to invest in lowering barriers, either financially, socio-culturally or with regard to the integration in the club. In the next section, we use this information as the dependent variable, to explore what predicts the concrete efforts of sports clubs.

Predictors of Nonprofit Sports Club’s Efforts

The second objective of this study is to identify predictors of the clubs’ efforts to lower barriers, more particularly for people in poverty. Table 6 presents the results of the bivariate analyses. The four sets of factors considered in this study—the organisational capacity of the sports club, their dependence on municipal subsidies, the presence of underprivileged people, and the competitive and/or recreational nature of the sports practice—are all associated with the club‘s efforts towards people in poverty. All predictors in the analysis show a statistically significant association with the dependent variable, with the exception of the variable concerning the availability of trainers, as well as the variable on sufficient accommodation.

Table 6 Results of the bivariate analyses for the percentage of sports clubs lowering barriers for people in poverty (N = 217)
Full size table

With regard to the organisational capacity of the club, sports clubs with larger financial resources are overrepresented among the sports clubs who make inclusionary efforts. Looking at the human resources capacity, and more particularly the presence of board members and volunteers (for refereeing, organising additional activities, maintaining infrastructure, cafeteria), the direction of the association contrasts with our expectations: sports clubs who state having sufficient board members and/or volunteers reported efforts for lowering barriers less frequently as compared to their counterparts. The structural capacity matters as well, since we observe that lowering barriers for people in poverty is more widespread among clubs that have a policy plan, as compared to their counterparts. Cooperation with local sports authorities is associated with the inclusionary efforts of the club as well. Nonprofit sports clubs with a cooperation with local sports authorities report lowering barriers for people in poverty far more often, as compared to their counterparts. Sports clubs with both recreational and competitive sports state to invest in people in poverty more often as compared to sports clubs which offer recreational sports only.

The strongest difference, however, is related to the presence or absence of underprivileged people in the club. As expected, sports clubs in which the representative indicates that underprivileged people are participating in the club are far more likely to make inclusionary efforts.

In order to ‘isolate’ the effects of the different variables and to verify the stability of these results when keeping other factors constant, a binary logistic regression has been conducted (model 1), presented in Table 7.

Table 7 Results of the binary logistic regression for the percentage of sports clubs lowering barriers for people in poverty, expressed in odds ratios (Model 1)
Full size table

In line with the findings of the bivariate analysis, one variable stands out as a very strong predictor in the regression results, i.e. the presence of underprivileged people. When controlling for other variables, the odds of making inclusionary efforts are estimated to be almost seven times higher (point estimate of 6.5) in clubs where underprivileged people are participating (as perceived by the club representative).

In addition to the impact of the social composition of the club, the organisational capacity also affects the club’s efforts towards people in poverty, though not fully as expected. It is mainly the presence of sufficient board members, one of the variables linked to human resources capacity, which jumps out as a predictor. Yet, as was already suggested by the results of the bivariate analysis, contrary to our initial expectations, the results point towards a lower likelihood of additional efforts for people in poverty in sports clubs with sufficient board members. In the logistic regression, controlling for other variables, the point estimates for the effect of having enough volunteers are below 1 (suggesting a lower likelihood) as well, but the effect is not statistically significant. We cannot exclude, however, that this is related to a lack of statistical power, given a relatively low number of cases (217) compared to the number of variables in the analysis.

The results indicate that the structural capacity of the club determines the club’s efforts towards lowering barriers for people in poverty as well. More precisely, keeping other factors constant, clubs who have a policy plan are more likely to invest in people in poverty (point estimate of 2.3) as compared to their counterparts.

Based on the results of the logistic regression presented in Table 7, there is no evidence that the financial capacity has an impact on the club’s efforts in terms of lowering barriers for people in poverty, nor do we find any evidence of an impact of having sufficient accommodation, nor of differences related to competitive versus recreational sports practice. Similarly, we do not find any evidence of an impact of the cooperation with local sports authorities, nor of the dependence of local subsidies. In analogy with the impact of volunteers, however, it should be noted that the estimated coefficients for the variables related to the local government are positive and rather large. In other words, here too, a lack of statistical power could possibly prevent an effect to be apparent.

In order to evaluate the stability of the results between the different efforts in terms of lowering barriers, Table 8 presents the results of the regression analyses for the separate items, i.e. financial barriers (model 2), socio-cultural barriers (model 3) and facilitating integration (model 4). Keeping other variables constant, a neutral answer on the question whether there is sufficient accommodation is related with a lower likelihood of lowering barriers for people in poverty in all three models, while this coefficient is not statistically significant in model 1 (Table 7). Yet, overall, the results are similar between items, and in line with the results presented in Table 7. Regarding the effect of having a policy plan, with a p value of 0.051 in model 2, the result is comparable to model 3 and model 4.

Table 8 Results of the binary logistic regression for the percentage of sports clubs lowering financial barriers (Model 2), socio-cultural barriers (Model 3), and facilitating integration (Model 4) of people in poverty, expressed in odds ratios
Full size table

Discussion

In a context of heightened expectations towards nonprofit sports clubs in terms of achieving broader welfare-related government objectives, a first aim of this study was to explore the attitude of nonprofit sports clubs towards fulfilling a role in social inclusion in sport. From the findings we observe that an overwhelming majority of the club representatives (94 %) agree with the statement that everybody is welcome at their club, regardless of a person’s social, cultural or ethnic background. However, policies and initiatives to accomplish social inclusion are not present to the same extent. This is in line with Rulofs (2012), who states that sports clubs will not formally exclude groups of people, but that this does not imply that clubs are open to everyone in practice, as exclusion mechanisms tend to be subtle in nature (Elling and Claringbould 2005; Rulofs 2012). In addition, clubs are likely to be reluctant to deal with change, including with regard to diversity (Rulofs 2012). In our study, we have found that dealing with diversity and facilitating social inclusion are structurally anchored in the functioning of the sports clubs only to a very small extent. Yet, while the formal integration of the inclusion objective is limited, approximately half of the sports club representatives agree that it is the task of the club to encourage diversity among the members and to try to accomplish that disadvantaged groups in society also take part in the club. Still, we also find that almost four out of ten club representatives do not report any efforts for lowering barriers for disadvantaged groups. The findings indicate that, in spite of isomorphic processes, there is a heterogeneity among sports clubs. If clubs are influenced by their institutional context, social inclusion is not yet an overly determining value in the field and indeed gets overshadowed (see Fahlén and Karp 2010; Skille 2011; Stenling and Fahlén 2009).

The second aim of this study was to predict the presence or absence of efforts of sports clubs for lowering barriers to participation, more particularly for people in poverty, who can be considered as a disadvantaged group with regard to sports participation. Based on the findings in the literature that human resources are the most pressing problems of clubs (Seippel 2004), and that volunteers felt their club did not have the capacity to deliver policy goals (see f.i. Harris et al. 2009), we expected that clubs with more human resources would be more likely to make inclusionary efforts. Yet, the results indicate the opposite, i.e. clubs with sufficient board members are less likely to lower barriers for people in poverty. The same may hold for having enough volunteers, though the evidence is not conclusive in this respect, and further research is needed to empirically verify this. Yet, the findings indicate that human resources capacity—or a lack of it—is not the key concern here. Interpreting these findings, it seems likely that the main factor is rather to be found in the conclusions of Nagel (2008) and Reid (2012), amongst others, stating that clubs are first and foremost oriented towards their members and their members’ interests. Exactly because there are sufficient board members, there is no obligation to open up and look outside the customary boundaries of the club. Board members in a club are often members that have a long experience in club sport, and who identify strongly with the club and its values. It therefore seems probable that in clubs with enough board members, the culture of similarity, as described by Doherty and Chelladurai (1999) will be stronger. In other words, interpreting the findings, we assume that in sports clubs with a stronger ‘inner circle’, there is more focus on similarity, and hence fewer externally oriented investments. More generally, the findings indicate that the organisational capacity of the sports clubs might not be as problematic as suggested in the literature, in the sense that larger financial resources nor sufficient availability of volunteers or board members, nor a large number of members, significantly increases the odds of taking measures for lowering barriers for people living in poverty, at least not when controlling for other factors. We were not able to control, however, for the motivation of the volunteers or board members. In other words, whereas the number of volunteers does not seem to affect the efforts made by the sports clubs, it might still be the case that promoting social inclusion is hampered by being too distinct from the main motivation of the volunteers, as mentioned by Coalter (2007b) and Harris et al. (2009), amongst others. Similarly, also the educational level, and more generally, the profile of the board members and (other) volunteers (in terms of capacities, preferences, interests), may be related to the efforts to facilitate participation. Yet, we could not empirically verify this in the analyses.

Another dimension of organisational capacity is found to be of influence, more particularly of the presence of a policy plan, i.e. a document in which the strategic goals of the club are outlined. This suggests that clubs in which the board takes a more professional, managerial approach are more likely to make inclusionary efforts. In the literature, it is argued that the diversity of sports clubs has to be taken into account when formulating wider expectations towards clubs. Clubs with a strong structural capacity in terms of planning and development are likely to be better prepared to strive for social inclusion.

Skille (2011) has argued that with competitiveness being the dominant convention in sports clubs, the possibilities for sports clubs to realise social goods are constrained. Yet, no evidence was found that sports clubs offering competitive sports would be less likely to pay attention to lowering barriers for people in poverty. This implies there is no difference in the effort made by the sports clubs, by level of practice. However, it might still be the case that competitive sports clubs are less (or more) likely to attract people from socially disadvantaged groups, for example by their organisational culture. In addition, it should be mentioned that only 8 % of the clubs in our sample were mainly competitive.

The results show that clubs in which also underprivileged people take part are much more likely to make inclusionary efforts towards people in poverty, and the size of the effect is very large. The analysis does not allow to identify the direction of the effect. The board of the club may take measures to lower barriers because they are aware that some members are in a difficult situation (and that therefore, potential new members might be too). Reversely, it might also be the case that clubs where initiatives are taken to lower barriers are more accessible for all, with the presence of underprivileged people as a result. It seems most likely that both are true, and we are dealing with a bidirectional effect. A somewhat striking element in this observation, however, is the size of the effect, even when controlling for the organisational capacity of the club. The large overlap between clubs that are actively lowering barriers for people in poverty, and clubs where some of the members are underprivileged (according to the club representative), suggests that to a certain extent, there is a dichotomy in the club sector: on the one hand, there are ‘social’, or socially diverse clubs, which try to be open to all and have members of different social layers. On the other hand, there are clubs in which social inclusion is not on the agenda, and poverty and lack of opportunities (seemingly) remain ‘distant’ problems.

Conclusion

This research investigated the role of sports clubs regarding the inclusion of socially disadvantaged groups. The study was conducted in Flanders (Belgium), based on a survey among sports clubs (Flemish Sports Club Panel) from 2012. Our study indicates that approximately half of the club representatives consider it as the task of their club to encourage diversity among the members and to include disadvantaged groups in the club. Specifically with regard to people in poverty, the findings indicate a similar share of clubs, i.e. about one out of two clubs, currently do not make any concrete efforts for lowering barriers to participation. The results indicate a cleavage in the club sector. On the one hand, there are clubs who are fully ‘on board’ when it comes to fulfilling wider social goals, more particularly social inclusion. In other clubs, on the other hand, social issues like poverty and social disadvantage still seem a distant problem.

The study has some limitations. First, the large number of missing values on the dependent variable (36 %) may have influenced the results. Yet, analyses revealed that the missing values are unrelated to the respondents’ opinions with regard to diversity, which suggests it is unlikely that the results will be seriously biased. The length of the survey as well as the extensiveness of the question (as part of a wider table) at least partially account for the number of missing values. Second, socially desirable responding may have had an impact on the results as well. Third, as this study is explorative, it is possible that some important determinants were not included in this study. Fourth, and along the same line, no information on ‘organisational culture’ could be included in the analysis. The same holds for the profile of volunteers and board members. Further research is needed to offer additional clarification in this regard. Organisational culture may however be hard to grasp in a quantitative variable. In this study, we have focused on structural characteristics. To investigate the impact of a club’s culture, qualitative research is more suitable. The results of this study were not fully conclusive with regard to the impact of the number of volunteers as well as the effect of a partnership (financial and otherwise) with local sports authorities. Further research is needed to corroborate (or contradict) the findings of this study in this regard.

Despite its limitations, the current study has contributed to the present knowledge in the field of sports policy and voluntary sector research, and the results are informative for scholars as well as policy makers. The research shows that it is not so much the ‘capacity’ of the clubs in terms of quantity of resources which is likely to hamper taking inclusionary measures for people in poverty. It seems rather a cultural matter, as having sufficient board members deters clubs from taking measures favouring the inclusion of people in poverty. This study has also indicated that (the awareness of) the presence of underprivileged people is very strongly associated with lowering barriers for people in poverty. Both findings suggest that, if clubs are expected to contribute to tackling inequalities in organised sports, instilling awareness in clubs about poverty and social inequality is likely to be a necessary step. Sports authorities and sports federations still have an important role to play in supporting and encouraging nonprofit voluntary sports clubs in this regard. Social diversity is not a ‘goal’ in itself, but a social reality that cannot be ignored.