Individual and contextual determinants of stable volunteering in sport clubs

Abstract

This article addresses factors that influence voluntary sport club (VSC) members’ loyalty to voluntary engagement. The question asked is an issue of VSC volunteers’ commitment; whether they decide to quit or continue their engagement. A multilevel approach was used that considered both individual characteristics of volunteers and corresponding contextual features of VSCs to analyse members’ voluntary commitment. Different multilevel models were estimated in a sample of 477 volunteers in 26 Swiss and German VSCs. Results indicated that members’ stable voluntary activity is not just an outcome of individual characteristics such as having children belonging to the club, strong identification with their club, positively perceived (collective) solidarity and job satisfaction. In addition to these factors, the findings confirm the significance of the contextual level. Stable volunteering appears to be more probable in rural VSCs and clubs that value conviviality. Surprisingly, the results reveal that specific measures to promote volunteering have no significant effect on voluntary commitment in VSCs.

Keywords

contextual features multilevel analysis social action voluntary commitment voluntary sport clubs

Introduction

Although many people undertake volunteer work in voluntary sport clubs (VSCs) and their engagement can be characterized as stable, many other volunteers occasionally or often think about terminating their voluntary engagement. When recent VSC monitoring studies in various European countries asked club representatives to describe club-specific problems, the clubs reported that stable voluntary engagement cannot be seen as a matter of course, and binding board members and coaches to formal positions within the club appears to be one of the most important challenges (see the overview in the handbook Sport clubs across Europe edited by Breuer et al., 2015b). However, stable voluntary engagement in VSCs is crucial, as tasks are not usually formalized and specific knowledge can be lost when people are replaced. Short-term or high fluctuation rates of volunteers may be inefficient in processes of service provision.

Against this background it is hardly surprising that the binding of volunteers is one of the most serious problems facing many VSCs. Nonetheless, it should be considered that the purported decline in the willingness to volunteer has long been the subject of a controversial debate on the ‘crisis of volunteering’ in VSCs. Research also indicates that this purported crisis has not led to any reduction in the sport services offered by VSCs in the past (Anthes, 2009; Pitsch and Emrich, 1997, 1999). In particular, Pitsch (1999) has used the example of voluntary engagement in VSCs to discuss the problem of ideological influences on empirical research, and emphasizes deficits in the validity of measurements of shortages or crises in voluntary engagement. Nevertheless, sport club research does provide an ambivalent picture of voluntary engagement: club monitoring reports reveal that some VSCs have few problems gaining the stable voluntary engagement of their members, whereas others have considerable instability and many vacancies (Breuer et al., 2015a; Lamprecht et al., 2012). Hence, it can be assumed that stable voluntary activities of club members may also depend on the (distinctive) contextual factors within each club (e.g. its size, services, culture or strategic orientation). Because voluntary commitment usually occurs within a specific organizational context, the unique characteristics of the organization itself should also be considered (Studer and Von Schnurbein, 2013). Nonetheless, it can be assumed that the risk of terminating volunteering among club members is not just an individual problem. Different contextual aspects such as organizational conditions in VSCs may lead to differences in collectively shared action orientations, and this may be accompanied by individual differences in motivational characteristics or perceptions of volunteering that influence voluntary behavior in terms of stability of engagement or time donated to volunteering. Thus, focusing on the organizational settings affecting volunteers sheds light on the ‘meso-level’ between the above-sketched micro-level of motives, sociodemographic characteristics and personal dispositions and the macro-level of societal values, government policies and social capital affecting volunteering (e.g. Haski-Leventhal et al., 2009; Hustinx and Meijs, 2011).

Due to the importance of volunteers as a fundamental resource to produce sporting and social services in VSCs, the aim of the following study is to analyze both individual and contextual factors that determine the tendency to continue or to terminate voluntary engagement in VSCs.

Literature review

A review of the literature reveals numerous studies that have analyzed the personal characteristics of VSC volunteers, such as their motives, personal dispositions and socio-economic characteristics (see Wicker and Hallmann, 2013, for a summary).

Going beyond socio-economic variables such as age, income or education level, several studies have performed theoretical and/or empirical analyses of the motives associated with voluntary engagement from a variety of perspectives and in a variety of different sport contexts (e.g. Braun, 2003; Emrich et al., 2014; Farrell et al., 1998; Flatau et al., 2014; Kim et al., 2013; Strigas and Jackson, 2003). These motives reflect the values and goals individuals attribute to themselves and associate with voluntary activity. In general, results from different studies on volunteer motivation indicate that motives are located on a continuum between an altruistic/selflessness orientation (‘wanting to help others’) and an egoistic orientation (‘pursuing one’s own interests’) (Braun, 2003). The majority of studies support the notion that volunteering motivations are multifaceted; that is, they occur in combination with each other rather than in isolation (Rehberg, 2005). Because volunteers do differ in terms of their motivation for volunteering, one can find heterogeneous motivation-based profiles of voluntary engagement (Braun, 2003; Dolnicar and Randle, 2007). These diverging bundles of motives also help to explain why members engage in completely different activities and task domains (Clary and Snyder, 1999). Furthermore, volunteers’ motives have been linked not only to demographics such as gender (Skirstad and Hanstad, 2013) or age (Hardin et al., 2007; Okun and Schultz, 2003), but also to the frequency of past volunteering in order to analyze the evolution of motivation over a volunteering career and the life cycle (Emrich and Pierdzioch, 2014; Haski-Leventhal and Bargal, 2008).

Although motives are a powerful factor when it comes to explaining why individuals engage voluntarily in VSCs, the issue of voluntary commitment is also linked closely to other aspects. Current studies analyze how other factors such as volunteers’ satisfaction or identification with the club relate to the intention to quit voluntary engagement in VSCs (Schlesinger et al., 2013), time commitment (Burgham and Downward, 2005; Hallmann, 2015) and long-term volunteering at sporting events (Kristiansen et al., 2015).

Whereas there is an extensive body of studies on individual-level determinants, there has been only limited research on how contextual aspects such as organizational conditions and structures influence volunteering. A systematic literature review by Studer and Von Schnurbein (2013) emphasizes the relevance of the organizational context. It argues that the practices and instruments of volunteer management, and, even more strongly, the organizational attitudes towards volunteers as well as an organization’s embedded values co-determined by social processes (integration and production of meaning) are crucial factors affecting volunteers. The review also addresses structural features such as volunteer management capacity that limit the action space of volunteers and volunteer coordination. Concerning VSCs – as a specific sector of voluntary engagement – only a limited number of studies have focused on the organizational aspects associated with volunteering. First, comparisons in different voluntary settings (e.g. German Red Cross vs. German soccer clubs) reveal differences in voluntary motivation. Membership of the German Red Cross can be viewed as an ‘altruistic’ resource pooling in which members produce public goods mainly for non-members. Therefore, private consumption motives do not play a major role for voluntary supply (Emrich and Pierdzioch, 2015). However, in VSCs as interest organizations whose members produce goods for members, strong evidence is found for the private good consumption model (Hämmerle et al., 2014). Therefore, the taxonomy of altruistic versus egoistic resource pooling should deliver a more detailed understanding of motivational aspects in volunteering.

Further analyses are based mostly on data from comparative structural analyses in VSCs and reveal inconsistent findings. They show some effects of club size: VSCs with fewer members and a high share of members participating in social events have less difficulty in recruiting and retaining volunteers (Breuer and Wicker, 2010; Nagel, 2006; Scheerder and Vos, 2010). Furthermore, VSCs with high annual per capita revenues and no owned facilities experience smaller problems in recruiting and retaining volunteers (Wicker and Breuer, 2013). Regarding organizational objectives, it becomes clear that VSCs that can be characterized more as solidarity-oriented communities that set value on conviviality reveal a stronger willingness for members to volunteer (Wicker and Breuer, 2013). Vice versa, the level of volunteering in more service-oriented clubs providing a wide range of courses (also for non-members) is lower (Nagel et al., 2004). However, findings on Norwegian VSCs reveal that the level of commercialization was already significant among clubs, and that increasing commercial resources did not necessarily hinder or crowd out voluntary work (Enjolras, 2002). Regarding the influence of volunteer management practices, particularly Cuskelly et al. (2006) have investigated the efficacy of management practices in retaining volunteers in rugby clubs. Their results reveal that planning, training and support are associated with fewer problems in retaining volunteers. Alongside the positive effects of some volunteer management practices, however, VSCs that are guided by the idea of a service delivery organization and thus restrain individual choice for action through having more formalized work processes are less attractive for voluntary engagement (Hoeber, 2010; Nichols and James, 2008).

Despite the number of available studies, the state of research on voluntary commitment in VSCs is unsatisfactory in several respects. First, only a few studies relate the characteristics of volunteers to their club membership. Most pay no attention to how experiences as a club member within the club determine their decision to volunteer. Existing panel data or voluntary surveys seem to be rather unsuitable for analyzing such relations, because they are generally too aggregated. There is a far stronger need to gather data focusing on a club as a specific social context in order to obtain the most detailed information possible on different aspects of the relation between members and their club. Second, the available studies try to explain volunteering in VSCs separately on either the individual or the contextual level. Few analyses consistently link together individual data from volunteers with the corresponding contextual conditions in their VSC, as has been performed in other voluntary sectors (Bühlmann and Freitag, 2007; Rotolo and Wilson, 2012). Although there have been calls for linkages between individual and corresponding contextual data within sport organization research for some time (Nagel, 2007; Wicker and Hallmann, 2013), few studies in sport club research have applied multilevel procedures from an empirical perspective (of willingness to volunteer: Schlesinger and Nagel, 2013; member commitment: Schlesinger and Nagel, 2015). Consequently, we still know little about the influences of factors from different levels on voluntary commitment in VSCs, or about their interplay. As a result, studies fail to meaningfully explain why and how different factors exert an influence.

Theoretical framework

When dealing with questions of individual action in social contexts such as volunteering in VSCs, analyses are often reduced to individual-oriented research. However, the social embedding of individual behavior should be considered more carefully (Coleman, 1990; Esser, 1999). Furthermore, it should be taken into account that the decision to terminate a voluntary activity is different to the decision of willingness to volunteer (starting is different from quitting); therefore, other aspects may influence individuals’ decision. By considering voluntary commitment in VSCs as a type of specific of social action (Esser, 1999), the underlying theoretical model is framed by a multilevel perspective (see Wicker and Hallmann, 2013). The multilevel framework incorporates the fact that a volunteer is socially embedded (nested) within a specific VSC that encompasses both individual factors and contextual conditions.

Individual level

A number of studies have shown that ‘true’ altruism in general has a subordinate role in organized voluntary work. According to the ‘logic of selection’ (Esser, 1999), voluntary work describes – in the sense of the utility-expectation theory (Becker, 1976) – an exchange of time and effort for different rewards compared to those of work. The utility of voluntary work results from individual expectations and appraisal of conditions of volunteering alongside the incentive structures in the club (Downward et al., 2009; Emrich et al., 2014). Such rewards for voluntary engagement may be, for example, social recognition, an enhanced reputation, gaining social contacts or returns of social appreciation from other club members (Erlinghagen, 2003; Flatau, 2009). These rewards contribute to the individual utility of voluntary engagement. It is important to note that the reward must meet a certain standard (fulfill a particular expectation), or the exchange may come to a halt. Declining voluntary commitment can be seen as the result of a negative outcome of cost–utility considerations due to diverging individual expectations and organizational conditions and incentives (Emrich et al., 2012). If the volunteer experience does not deliver these returns, individuals are likely to leave in pursuit of more gratifying leisure activities (Downward et al., 2009). Therefore, individual expectations and motives regarding volunteering and whether these are met are of central importance (Clary and Snyder, 1999), as volunteers may adapt their expectations to the contextual features of the VSC only to a certain point. From the individual’s perspective, voluntary engagement is neither an obligation nor a need, as it is also possible to be a member of a VSC and participate in its services (club goods) without directly helping to produce the club goods.

Hence, volunteers develop their own particular needs and expectations regarding their volunteer activity (Chelladurai, 2006), as individual expectations regarding voluntary work are integrated into a broader context of social conditions and social behavior (Schlesinger and Nagel, 2011). Factors such as age and gender influence individual preferences, and motives also need to be considered (see Emrich and Pierdzioch, 2014), as they lead to differences in the extent to which volunteering promotes their utility. Furthermore, many aspects of utility have to be conceived as experience-related, i.e. acquired only after engaging in voluntary activity for some time. Thus the expectation towards volunteering can vary depending on how much and for how long a person volunteers and for which job functions they volunteer (Ivancevich and Matteson, 2002). The extent of the individual utility of volunteering for voluntary engagement in VSCs is also associated with the intensity of club membership and how committed they are to their club (Cuskelly and Boag, 2001; Flatau, 2009). In this context, volunteers may well differ in the extent of their collective solidarity and identity with their club in line with club-specific socialization and integration processes – the social process through which individuals gradually acquire club-specific norms and a growing sense of belonging to a VSC (Flatau, 2009; Haski-Leventhal and Cnaan, 2009).

Contextual level

The following section considers the contextual features of VSCs that may affect the individual decision of club members’ according to their voluntary engagement. However, the notion of context should not be understood as descriptive but as analytical, in the sense that the package of features characterizing a context depends on which research question is being addressed (Boudon, 2014).

Firstly, in more general terms, VSCs can be characterized in terms of their specific social structure as interest communities with an organizational logic based on self-organization and (egoistic) resource pooling (cf. Coleman, 1974). The basic idea of VSCs is to produce certain club goods such as sports and social services with the help of volunteer services, and to provide these goods exclusively for the utility and interests of their members. Club goods are captured as ownership–membership arrangements that are different from public or private goods (Buchanan, 1965). This ‘inside-oriented ethos’ (Schimank, 2005) clearly distinguishes VSCs from other types of (sports) organizations where volunteer work helps to produce goods or services for persons whom the volunteer usually does not know (e.g. The Red Cross, Olympic Games or other events). The production of the club goods in VSCs depends on actions based on solidarity among club members (Heinemann, 2004). Therefore, service provision relies on members who are willing to volunteer for a longer term because high fluctuation rates may be inefficient. This results in a specific kind of motivation of club members to invest time voluntarily that simultaneously validates the private good consumption model of VSC volunteers (Emrich et al., 2012; Hämmerle et al., 2014).

Beside these general aspects of VSCs as interest communities, every club can be characterized by specific contextual conditions that separate them from other clubs. VSCs as communities of shared interests are guided substantially by (distinctive) contextual aspects such as their specific history, culture and norms, and they develop guiding programmatic principles, which in turn influence their actions (Thiel and Mayer, 2009), and that also determine the individual expectations of volunteers (Peters-Davis et al., 2001). Volunteers are rooted within their club, and volunteers in each club share specific structural and cultural similarities that are unique to their club and may not be shared with other clubs (Schlesinger and Nagel, 2015).

It would appear that club size may be a relevant factor. In larger clubs, members’ interests are more heterogeneous, which may negatively affect volunteering (Wicker and Breuer, 2013). Furthermore, the size of an organization controls the relationship between resource investment and expected benefit in the creation of club assets (Buchanan, 1965). Hence, the size of the club can have opposing effects on volunteering: although perceived social appreciation by other members should increase with the number of members (Erlinghagen, 2003), larger clubs can reduce the social obligation to volunteer (decline in community character and growth in anonymity) and therefore encourage the free-rider problem (Heckathorn, 1989).

Furthermore, volunteering can be understood as a value-based activity, therefore both a club’s strategic orientation and the organizational values with respect to volunteering are crucial. The structurally and culturally embedded values and norms regarding volunteering in each specific club are essential elements that influence the ways in which members pool work donations (Emrich et al., 2015; Haski-Leventhal and Bargal, 2008), and they also determine the social appreciation of voluntary activity. Accordingly, two types of VSCs can be distinguished that represent the poles of a continuum: at one end, growth orientation, and on the other, social closure (Nichols et al., 2012; Schimank, 2005). In the first case, the aim is to provide sports for everyone, and new services and offers for different target groups are continually being developed for this purpose. This increasingly erodes the attachment between members and their club, thereby reducing the obligations to solidarity and also may destabilize voluntary commitment (Horch, 1998). On the other hand, however, are VSCs based on their traditional content, meaning that they focus strongly on their role as a community based on mutual solidarity (Nagel, 2006). Although solidarity-based obligations between a VSC and its members are not a binding contract, the members’ interpretation of agreements made with their club may influence perceptions of obligation and loyalty (Heinemann, 2004). Thus, the more strongly the importance of volunteering is internalized in the club, the greater the individual utility of volunteering and the higher the costs of non-volunteering.

The status of voluntary engagement within a club manifests clearly in measures to promote volunteering in the club along with the established practices of giving symbolic or material rewards (Cuskelly et al., 2006; Haski-Leventhal and Cnaan, 2009). However, a high level of formalization of working conditions that directly constrain the action space of volunteers may reduce the attractive utility for volunteers (Stirling et al., 2011). Even though it is generally held that low levels of bureaucracy and formalization contribute to the satisfaction of volunteers, Musick and Wilson (2008) have argued that very low levels could alienate them. Encouraging greater formality of VSCs is unlikely to align with the understanding, values or motives of sport clubs’ volunteers (Nichols and James, 2008). A key incentive for voluntary engagement in a VSC consists of not having to fulfill tasks strictly in line with the requirements of efficiency criteria. Furthermore, the incentive structures in VSCs should also be important. Although financial incentives are only generally significant for paid workers, there may well be other material incentives (‘fringe benefits’) that are important for volunteers. In the voluntary context, non-material gratification and recognition (e.g. volunteer events and letters of thanks) may be particularly relevant (Heinemann, 2004). However, external incentives provided by the club sometimes have counterproductive effects, as extrinsic incentives can have so-called displacement effects that undermine (intrinsic) values and morals of volunteering (see Frey, 1997).

The question of volunteering is also closely related to the alternative provisions that are available. Members of urban VSCs have markedly broader and more attractive leisure-time alternatives at their disposal. Broader and more attractive leisure-time alternatives make it increasingly less profitable to invest limited time in (temporally inflexible) voluntary work (Downward et al., 2009; Heinemann, 1995). Moreover, it is probable that rural VSCs continue to have a stronger community character due to their greater proximal value as social institutions (Tönnies, 1963). On the one hand, this may raise the value of the social appreciation of voluntary engagement for members. On the other hand, the lower anonymity and the accompanying possibilities of sanctioning non-commitment (free-riding becomes more costly) lead to the expectation of higher voluntary commitment (Bühlmann and Freitag, 2007).

Method

Data collection

The multilevel framework for analyzing the relationship between contextual conditions of VSCs and the individual actions of volunteers is characterized primarily by the combination of member- and club-related data. As it is desirable to have a large variety of contextual conditions at the club-level, the sampling procedure was designed to include different types of conditions in VSCs. Relevant selection criteria were the number of members and divisions, personnel structure, and types of sport. Club-specific data were obtained with a paper-and-pencil questionnaire by club managers (presidents and technical directors). The focus of this study is on whether club volunteers decide to terminate or continue volunteering. Accordingly, only club members who have a formal position (e.g. as coach, treasurer or secretary) in their club for a longer time were considered. Other members who only voluntarily engaged sporadically (e.g. short-term tasks at club events) were not considered. Individual-level data from volunteers with formal positions in their club were collected through an online survey.

The combination of individual and contextual data by multilevel analysis requires large sample sizes (Hox, 2002: 174). Having only few cases at the individual level only allows for limited conclusions, and ‘sample sizes less than 10 should be regarded with suspicion’ (Snijders, 2003: 676). Therefore, all VSCs delivering fewer than 10 observations on the individual level (volunteers) were dropped from the analysis. The resulting 26 VSCs with 477 volunteers on the individual level were entered for further analysis. Our analyses draw on data from 20 Swiss and 6 German VSCs. While belonging to different sport systems (with different laws and forms of subsidies), both Swiss and German VSCs can be integrated into this study as comparisons reveal a great deal of contextual similarity and overlap in the challenges they face – as identified by Wicker et al. (2014). Greater differences exist only in club size: Switzerland has more small VSCs. We integrated German VSCs to gain a higher variety of club sizes as a proxy for the degree of homogeneity or heterogeneity in the interests within a club.

The extent to which the sample is representative should be noted. A possible source of sampling bias is the failure of the study to entail a representative description of volunteering in VSCs. We were more interested in analyzing specific relations and variables in the individual decision-making processes.

Measures

The risk of terminating voluntary engagement was conceptualized as a decision-making act between the tendency for members either to continue or to terminate volunteering in their VSC (‘How often in the past few months have you felt like quitting your voluntary engagement for your sport club?’) on a 5-point scale ranging from 1 (often) to 5 (never). A total of 48.0% reported never; 18.4%, yes, but only occasionally; 17.8%, sometimes; and 15.8%, frequently and often combined. Hence, approximately two-thirds of volunteers could be characterized as stable and about one-third as unstable. Although the majority of those thinking about terminating their volunteering may well not actually do so, the risk of no longer volunteering was probably higher among those who had already entered into such a decision-making process.

The independent variables on the individual level were operationalized as follows: in a first step, individual preferences for (the utility of) volunteering is associated with satisfaction of their volunteering expectations. Volunteer job satisfaction is achieved when the expectations regarding working conditions are met (Chelladurai, 2006; Doherty, 2005). Therefore, a z-standardized index of volunteers’ job satisfaction was considered that covered five dimensions of work conditions in VSCs (task design, leadership, material incentives, recognition and support) identified in a prior study (Schlesinger et al., 2013). To operationalize the intensity of club membership, we included the variables membership duration, competition experiences, perceived collective solidarity and identification with the club. Measurements were based on elaborated measurements addressing aspects of the identification to VSCs (Cuskelly and Boag, 2001; Engelberg et al., 2012) and aspects of solidarity and collective interests in VSCs (Braun, 2003; Nagel, 2006).

The operationalization of independent variables on the organizational level was based on established concepts of other VSC studies (Cuskelly et al., 2006; Lamprecht et al., 2012; Wicker et al., 2014). Here we differentiated between situational variables (e.g. number of members and number of divisions), variables regarding supporting volunteering (e.g. strategic planning and incentive structure), and variables representing the strategic orientation of a club.

Table 1 gives an overview of the operationalization and descriptive statistics of all variables on both individual and organizational levels.

Table 1.

Variables for analyzing voluntary commitment, their operationalization and descriptive statistics.

Variable	Operationalization	Descriptive statistics
Dependent variable Voluntary commitment (Intention to quit voluntary activity)	‘How often have you felt like quitting your voluntary engagement for your sport club in the past few months?’ 1 (often) to 5 (never)	M = 3.7 (SD = 1.6); Yes, frequently = 9.4%; Yes, often = 6,4; Yes, sometimes = 17.8%; Yes, but only occasionally = 18.4%; never = 47.6%
Independent variables Individual level Gender Age Children belonging to club Competition experiences Duration of club membership Length of volunteering in the club Average time volunteered in the club Payments (up to 2000 CHF p.a.) Volunteer job satisfaction Identification with the club Perceived (collective) solidarity	Dummy; 1 = male Number of years of life (> 16 years) Dummy; 1 = yes (⩽ 16 years) Dummy; 1 = yes Number of membership years Number of years in voluntary work Number of hours per month for volunteering Dummy; 1 = yes Index from 5 dimensions (z-standardized) Index from 5 Items (z-standardized; Cronbach’s α = 0.83) Index from 4 Items (z-standardized; Cronbach’s α = 0.85)	Male = 70.1%; Female = 29.9% M = 40.6 (SD = 14.5) Yes = 17.2%; No = 82.8% Yes = 83.6%; No = 16.4% M = 19.3 (SD = 11.6) M = 11.7 (SD = 9.5) M = 16.6 (SD = 12.0) Yes = 33.7%; No = 66.3% M = 3.90 (SD = 0.9) M = 4.23 (SD = 0.7) M = 4.01 (SD = 0.7)
Club level Situational features Members Members² Divisions Community size, where the volunteer sport club (VSC) is local embedded	Number of club members Squared number of club members (members * members) Number of divisions with different sports 1 = rural; 2 = agglomeration; 3 = urban; 4 = city	M = 1,409.6 (SD = 2,152.6) M = 6,612,941.4 M = 9.0 (SD = 11.8) Rural = 10.3%; agglomeration = 29.1%; urban = 24.1%; city = 36.5%
Volunteer-related features of the VSC Problems with ‘volunteering’ Measures to promote volunteering^a Strategic planning of volunteering^b Immaterial/symbolic rewards Material incentives Paid staff within the club Strategic orientation of the club Supporting competitive sports Supporting grassroots sports Maintaining tradition Set value on conviviality Supporting external cooperation	1 = no problems to 3 = big problems Index of selected items (categorized: 1 = no further measures to 3 = specific measures) Index of selected items (categorized: 1 = no specific coordination to 3 = coordinated) 1 = no; 3 = strong 1 = no; 3 = strong Dummy; 1 = yes, paid staff 1 = not important to 5 = important 1 = not important to 5 = important 1 = not important to 5 = important 1 = not important to 5 = important 1 = not important to 5 = important	No problems = 27.9%; medium problems = 34.3%; big problems = 37.7% No further measures = 16.1%; scattered measures = 51.4%; specific measures = 32.5% No specific coordination = 19.3%; rather coordinated = 59.3%; coordinated = 21.3% No = 31.1%; partially = 39.1%; strong = 29.8% No = 50.6%; partially = 33.8%; strong = 15.6% Yes = 59.9%; No = 40.1% Not important = 5,3%; neither/nor = 11.1%; rather important = 62.8%; important = 20.8% Neither/nor = 13.7% rather important = 29.4%; important = 56.9% Not important = 3.2%; less important = 2.4%; neither/nor = 31.0%; rather important = 46.0%; important = 17.4% Less important = 7.1%; neither/nor = 27.7%; rather important = 27.9%; important = 37.3% Less important = 14.1; neither/nor = 23.1%; rather important = 49.6%; important = 13.2%

Notes:

Measures to promote voluntary engagement of club members include clearly defined volunteering responsibilities, specific publicity campaigns to recruit volunteers and keeping membership records. These were assessed on a 5-point scale ranging from 1 (not at all true) to 5 (completely true).

Items include strategic planning of volunteering such as job descriptions, volunteer coordinator, knowledge about skills, expectations of volunteers and volunteering as topic in mission statement. These were assessed on a 5-point scale ranging from 1 (not at all true) to 5 (completely true).

Data analysis

An analysis of contextual influences involves hierarchical data structures. Data are hierarchical when the observations on the individual level are not independent (Raudenbush and Bryk, 2002). This means that cases within a context unit (volunteer within a VSC) are more similar to each other than cases belonging to other, different context units (volunteers in other VSCs). Thus, it could be assumed that the voluntary commitment of individual club members varies less within VSCs than between VSCs. In this case, a so-called intra-class correlation (ICC) exists. In the case of hierarchical data, it is important that the data are measured at appropriate levels and this requires multilevel modeling data analysis techniques to avoid Type-1 errors (Peugh, 2010). The appropriate method for simultaneously assessing data on both individual and contextual levels is a multilevel analysis. The multilevel analysis guarantees the adequate handling of hierarchically structured data, as neither a pure aggregate nor a pure individual data analysis would deliver satisfactory results. A purely individual data analysis would not do justice to the inherent hierarchic structures (Raudenbush and Bryk, 2002). Aggregate data would lead to the risk of false conclusions (e.g. ecological fallacy; see Robinson, 1950) and be open to the criticism of neglecting the original data structure (Snijders and Bosker, 1999). Multilevel analysis is an appropriate method to gain a more exact picture of the effects of social contexts by developed theoretical models implemented by statistics. Therefore the use of multilevel analysis allows for more deep understanding of individual behavior or decisions within VSCs.

Due to limited sample size, we estimated only random-intercept models with the restricted maximum likelihood procedure. In random intercept models, the slopes are fixed (this means the same across different contexts). Furthermore, due to constraints in the database (insufficient numbers of clubs) we did not estimate other models such as random slope models and the cross level interaction of individual and contextual entities. In line with the logic of multilevel models, we followed the established successive analyses procedure (e.g. Bühlmann and Freitag, 2007).

Results

The initial assumption was that voluntary commitment of individual club members varies within VSCs less than between VSCs. Therefore as a first step, we estimated a random-intercept-only model to analyze the variance proportions of the dependent variables on the individual and contextual levels. This model provides information on ICCs, which are helpful in determining whether multilevel models are required in the first place. Because the intercept-only model contained no explanatory variables, the variance terms represent unexplained residual variance. If no variance in the dependent variable could be determined on the club level, contextual features of the clubs indicate no further variance and thus, multilevel analysis is not required (Hox, 2002). The size of the ICC was applied as a criterion to determine whether multilevel analysis was an adequate modelling strategy. The estimated random intercept-only model indicates that the variance of the individual-level residuals (Var r_ij) was 1.467 (see Table 2). The variance of the context-level residuals (Var u_0j) was 0.139. Table 2 displays the variance components for both levels followed by the ICC. The calculated ICC was 0.087. This indicates that 8.7% of the variance could be traced back to contextual differences between the clubs. The estimation of the random intercept-only model confirms that voluntary commitment varies not only between individuals, but also between the contextual features of the clubs. Based on the ICC, it could be concluded that the contextual level matters. This finding confirms that multilevel analysis is appropriate.

Table 2.

Individual and contextual determinants of voluntary commitment in voluntary sport clubs (random intercept, non-standardized coefficients).

	Random intercept only	Random intercept (individual)	Random intercept (individual+contextual)	Random intercept (individual+contextual)	Random intercept (individual+contextual)	Random intercept (full) model
Fixed part
Intercept	3.935^***	4.073^***	4.224^* **	4.297^***	4.083^***	4.251^***
Individual level
Gender (1 = male)		0.010	0.014	0.016	0.015	0.014
Age		−0.001	−0.001	−0.001	−0.001	−0.001
Children belonging to club (1 = yes)		0.346^**	0.331^**	0.334^**	0.326^**	0.323^**
Competition experiences (1 = yes)		0.109	0.112	0.108	0.112	0.111
Volunteer job satisfaction		0.258^***	0.240^***	0.247^***	0.244^***	0.240^***
Duration of club membership		−0.004	−0.004	−0.004	−0.004	−0.004
Length of volunteering in the club		−0.030^***	−0.031^***	−0.030^***	−0.031^***	−0.030^***
Average time volunteered in the club		0.003	0.002	0.003	0.003	0.002
Identification with the club		0.362^***	0.359^***	0.351^***	0.357^***	0.351^***
Perceived (collective) solidarity		0.337^***	0.338^***	0.340^***	0.330^***	0.338^***
Payments (1 = yes)		−0.063	−0.056	−0.058	−0.052	−0.060
Club level
Members		–	−0.001	–	–	–
Members²		–	0.000	–	–	–
Divisions		–	0.004	–	–	–
Community size		–	−0.127^*	–	–	−0.112^*
Supporting competitive sports		–	–	0.106^*	–	0.087
Supporting grassroots sports		–	–	0.020	–	–
Maintaining tradition		–	–	−0.031	–	–
Set value on conviviality		–	–	0.132^*	–	0.118^*
Supporting external cooperation		–	–	−0.073	–	–
Paid staff within the club (1 = yes)		–	–	–	−0.007	–
Measures to promote volunteering		–	–	–	0.080	–
Strategic planning of volunteerism		–	–	–	0.053	–
Immaterial/symbolic rewards		–	–	–	0.069	–
Material incentives		–	–	–	0.049	–
Random part
Variance individual level (Var r_ij)	1.467	1.054	1.011	0.984	0.996	0.981
Variance contextual level (Var u_0j)	0.139	0.103	0.091	0.083	0.097	0.079
Intra-class correlation coefficient (p)	0.087	0.089	0.083	0.078	0.088	0.075
Deviance (-2 log likelihood)	1563.6	1358.5	1335.4	1336.9	1349.4	1326.2

p ⩽ 0.05; ^**p ⩽ 0.01; ^***p ⩽ 0.001.

In a second step, we entered the individual differences in voluntary commitment at the individual level into the multilevel model while controlling the potential structural differences. The random-intercept model documents the estimation of all theoretically developed variables at the individual level (see Table 2). Within this model, the intercepts vary and the scores on the dependent variable for each individual observation are predicted by the intercept that varies across groups. Results showed a significant influence of five variables at the individual level. A strong identification with the club, a positively perceived collective solidarity, satisfaction with the club’s working conditions and children belonging to the club had a positive influence on voluntary commitment; whereas the length of volunteering had a negative effect. Other individual factors such as gender, age, duration of membership or competition experiences had no significant influence on voluntary commitment.

To analyze contextual features of VSCs that determine commitment of volunteers, we carried out further estimations by adding bundles of club-related contextual variables step by step into the model and examining its explanatory power (Table 2). Results indicated three significant factors that explain club-related differences in voluntary commitment. More rural clubs, clubs in which supporting competitive sports is important and clubs that place value on conviviality revealed more stable voluntary commitment independently from individual characteristics. From a modeling perspective, it is advisable to remove the non-significant independent variables in subsequent stages to improve the quality of the model (Hox, 2002). Therefore, we carried over the significant contextual factors into a multivariate random-intercept (full) model that confirms the influence of two contextual variables: stable voluntary commitment was more probable in rural VSCs and in clubs that explicitly set value on conviviality; and the other structural variable ‘achieving competitive sport’ was no longer significant in the random-intercept (full) model.

In regard to model fit: firstly we can see that the ICC decreases in relation to variables at the context level. This makes it clear that the contextual level matters, and includes variables to explain differences between clubs. The R² for the context level indicates that VSC variables explain 23% of the variation between the VSCs regarding voluntary commitment. The R² for the individual level reveals that the individual variables explain 28% of the variation between individuals regarding voluntary commitment in a VSC. Furthermore, comparison with the deviance values (-2 log likelihood) indicated that the consideration of contextual data improved the model valuation, meaning that the model was adapted to the empirical data to a higher degree.

Discussion

The aim of the current study is to analyze the relation between the clubs and their members as volunteers and to explain the main causes as to why they terminate their voluntary engagement. The multilevel analysis indicates that the voluntary commitment of VSC members is influenced by individual characteristics as well as corresponding to contextual features of the club. Results show that individual expectations regarding the club’s working conditions and whether these expectations are met were important. More satisfied volunteers exhibited a lower risk of terminating their volunteering as shown in previous studies (e.g. Kim et al., 2007; Schlesinger et al., 2013; Silverberg et al., 2001). Furthermore, the stability of voluntary engagement is associated with identification towards the club and the perceived obligation of solidarity. The results emphasize the importance of aspects such as solidarity when explaining individual social actions such as continuing or terminating volunteering in organizations where members produce goods for members. This is similar to findings reported by Emrich and Pierdzioch (2015) that identified the private good consumption model in VSCs. The commitment of a VSC is to bring out the simultaneous satisfaction of collective and individual interest. The results of the research show that the lack of social commitment of some entities means that the individual commitment of some of its members is also missing. Therefore, an increasing service- and growth-orientation of VSCs distance their members from the volunteerism tasks that enable the feeling of solidarity and usefulness. It becomes clear that within an interest organization voluntary engagement of club members will be easier to retain for the club’s own use (provision of services for club members) than when external (non-members) benefit from club services. Moreover, having children who belong to the club also had a positive influence on voluntary commitment. This underlines the plausibility of the finding that having children currently belonging to the VSC had a positive effect on volunteer activities and therefore promoted the likelihood of stable volunteering (Burgham and Downward, 2005). In contrast, the length of volunteering has a negative effect on voluntary commitment. This means, the longer volunteers actively volunteered in their club, the greater the risk of termination. This finding can be interpreted as follows: with the increasing duration of voluntary engagement, task-related obligations are no longer attractive enough. This can also be seen as a sign of saturation, in the sense that a volunteer has done enough for her or his VSC, and it is time to pass on the tasks to ‘other (younger) hands’.

As initially assumed, the findings confirm that that the contextual level matters, and that voluntary commitment varies between VSCs. This is underscored by the finding that the intention to quit voluntary engagement is less prominent in rural clubs. Clubs in rural areas have more stable voluntary engagement regardless of the individual characteristics of their members. This can be traced to the fact that members of urban VSCs have access to a broader range of attractive recreational as well social alternatives at their disposal, so that terminating voluntary engagement is less costly as pointed out by Heinemann (1995). Furthermore, there is a strong influence of the community compared to the association that stabilizes the obligations to solidarity and commitment to the club’s interests. These factors increase the value of the social appreciation of voluntary engagement in the club and make free-riding more costly. Furthermore, the findings indicated that VSCs that more explicitly support conviviality can promote the stability of voluntary engagement. This confirms findings from previous VSC reports indicating that clubs that support conviviality have smaller fluctuation rates regarding volunteerism (Lamprecht et al., 2012; Wicker and Breuer, 2013). It appears that this orientation of clubs is particularly important to the retention of volunteers, even though conviviality is often viewed as an obsolete feature. Hence, recommendations (e.g. by sport policy or federations) that VSCs should become more professional like modern service companies – without supporting sociability that strengthens the social and emotional commitment within the club – must take care that this does not lead to a further destabilization of voluntary commitment. Results also showed that club size has had no influence on volunteers’ decision to terminate their activity. This is because the formal organizational unit is not the focus of the volunteers’ commitment to the club but rather the unit of the sport group or the division works as a point of reference, like a ‘club within a club’ (e.g. Haski-Leventhal and Cnaan, 2009). Moreover, our study surprisingly revealed that specific measures and strategies to promote volunteering have no explanatory power on voluntary commitment to the club. This suggests that previous measures and strategies have either been less effective, or that a high level of formalization in VSCs might be less attractive for volunteers – as pointed out in previous studies (Hoeber, 2010; Stirling et al., 2011). Further reasons may well be methodological. For example, contextual variables may have been too unspecified, or the differences between clubs may not have been sufficiently large (their selection was too homogeneous).

Although the present study provides insights and starting points, there is a strong need for further theoretical considerations and empirical analyses due to the many open questions and to the limitations of the current findings. First, the present study focused only on the intention to terminate volunteering. Future studies should include former volunteers of VSCs such as dropouts during the previous months, and also distinguish between different types of volunteering. Second, due to homogeneous selection of VSCs, the contextual variables used in the analysis do not demonstrate large variances between clubs. Therefore, further studies should recruit larger samples, particularly at the context level. Additionally, more cases at the context level permit further analysis options such as random-slope models or cross-level interactions that would deliver a deeper understanding of individual behavior within contextual conditions. Researchers have assessed the necessary sample size at the context level and concluded that at least 30 cases are necessary to have correct standard estimates at the context level (Maas and Hox, 2004). Third, the study shows that a club’s characteristics may significantly impact on the individual’s decision to continue their voluntary activity. Nonetheless, future research should specify the contextual conditions of VSCs such as material and immaterial incentives or opportunities for voluntary engagement more precisely. Finally, among the contextual conditions, the specific understanding and valuation of volunteering in VSCs varies across different sports systems and this could have different effects on the decision to engage in volunteering. Further research in various countries is needed in order gain more broadly generalizable findings.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was financally supportet by the Swiss Federal Office of Sport (FOSPO).

References

Anthes

(2009) Strukturelle Entwicklungen von Sportvereinen: eine empirische Studie [Structural Developments of Sports Clubs: An Empirical Study]. Göttingen, Germany: Cuvillier.

Becker

(1976) The Economic Approach to Human Behavior. Chicago, IL: University of Chicago Press.

Boudon

(2014) What is context? Kölner Zeitschrift für Soziologie und Sozialpsychologie 66(Supplement 1): 17–45.

Braun

(2003) Leistungserstellung in freiwilligen Vereinigungen. Über ‘Gemeinschaftsarbeit’ und die ‘Krise des Ehrenamts’ [Providing services in voluntary associations: ‘Community work’ and the ‘volunteering crisis’]. In: Baur

Braun

(eds) Integrations-leistungen von Sportvereinen als Freiwilligenorganisationen. Aachen, Germany: Meyer & Meyer, pp.191–241.

Breuer

Wicker

(2010) Ehrenamt. Sportentwicklungsbericht 2009/2010-Analyse zur Situation der Sportvereine in Deutschland [Volunteering. Report on development in sport 2009/2010: The situation of sport clubs in Germany]. Cologne, Germany: Deutsche Sporthochschule Köln.

Breuer

Feiler

Wicker

(2015a) Sport clubs in Germany. In: Breuer

Hoeckman

Nagel

et al . (eds) Sport Clubs in Europe: A Cross-National Comparative Perspective. Heidelberg: Springer, pp.187–208.

Breuer

Hoeckman

Nagel

et al . (eds) (2015b) Sport Clubs in Europe: A Cross-National Comparative Perspective. Heidelberg, Germany: Springer.

Buchanan

(1965) An economic theory of clubs. Economica 32(125): 1–14.

Bühlmann

Freitag

(2007) Freiwilligentätigkeit als Sozialkapital. Eine empirische Analyse zu den Rahmenbedingungen bürgerschaftlichen Vereinsengagements [Volunteering as social capital. An empirical analysis toward framework of voluntary engagement in organisations]. Kölner Zeitschrift für Soziologie und Sozialpsychologie, Sonderheft 47, Sozialkapital. Grundlagen und Anwendungen 163–182.

10.

Burgham

Downward

(2005) Why volunteer, time to volunteer? A case study from swimming. Managing Leisure 10(2): 79–93.

11.

Chelladurai

(2006) Management of Human Resources in Sport and Recreation. 2nd ed. Champaign, IL: Human Kinetics Publishers.

12.

Clary

Snyder

(1999) The motivations to volunteer: Theoretical and practical considerations. Current Directions in Psychological Science 8(5): 156–159.

13.

Coleman

(1974) Power and the Structure of Society. New York, NY: W.W. Norton & Company.

14.

Coleman

(1990) Foundations of Social Theory. Cambridge, MA: Belknap.

15.

Cuskelly

Boag

(2001) Organisational commitment as a predictor of committee member turnover amongst volunteer sport administrators: Results of a time-lagged study. Sport Management Review 4(1): 65–86.

16.

Cuskelly

Taylor

Hoye

et al . (2006) Volunteer management practices and volunteer retention: A human resource management approach. Sport Management Review 9(2): 141–163.

17.

Doherty

(2005) Volunteer Management in Community Sport Clubs: A Study of Volunteers’ Perception. Toronto, ON, Canada: The Sport Alliance of Ontario.

18.

Dolnicar

Randle

(2007) What motivates which volunteers? Psychographic heterogeneity among volunteers in Australia. Voluntas 18(2): 135–155.

19.

Downward

Dawson

Dejonghe

(2009) Sports Economics: Theory, Evidence and Policy. Amsterdam, The Netherlands: Elsevier.

20.

Emrich

Pierdzioch

(2014) Die Motive Ehrenamtlicher im Sport: Eine Lebenszyklus-analyse [The motives of volunteers in sport: A life cycle analysis]. Spectrum der Sportwissenschaften 26(1): 4–24.

21.

Emrich

Pierdzioch

(2015) Testing economic models of volunteer labor supply: Some empirical evidence for the German Red Cross. Applied Economics 47(40): 4247–4259.

22.

Emrich

Pitsch

Flatau

(2014) Voluntary engagement in sports clubs: A behavioral model and some empirical evidence. International Review for the Sociology of Sport 49(2): 227–240.

23.

Engelberg

Zakus

Skinner

et al . (2012) Defining and measuring dimensionality and targets of the commitment of sport volunteers. Journal of Sport Management 26(2): 192–205.

24.

Enjolras

(2002) The commercialization of voluntary sport organizations in Norway. Nonprofit and Voluntary Sector Quarterly 31(3): 352–376.

25.

Erlinghagen

(2003) Die individuellen Erträge ehrenamtlicher Arbeit. Zur sozioökonomischen Theorie unentgeltlicher, haushaltsextern organisierter Produktion [Individual benefits of voluntary work. Toward socioeconomic theory of non-paid, household external organised production]. Kölner Zeitschrift für Soziologie und Sozialpsychologie 55(4): 737–757.

26.

Esser

(1999) Soziologie – Spezielle Grundlagen. Band 1: Situationslogik und Handeln [Sociology – Special Principles. Volume 1: Situational Logic and Action]. Frankfurt, Germany: Campus-Verlag.

27.

Farrell

Johnston

Twynam

(1998) Volunteer motivation, satisfaction, and management at an elite sporting competition. Journal of Sport Management 12: 288–300.

28.

Flatau

(2009) Zum Zusammenhang von Sozialisation und ehrenamtliche Mitarbeiter in Sportvereinen: Erste Überlegungen unter Anwendung der Rational-Choice-Theorie [Relation between socialization and volunteers in sports clubs: Preliminary ideas when applying rational choice theory]. Sport und Gesellschaft 6(3): 259–281.

29.

Flatau

Emrich

Pierdzioch

(2014) Einfluss unterschiedlicher Motive auf den zeitlichen Umfang ehrenamtlichen Engagements in Sportvereinen [Influence of different motives on the temporal scope of voluntary work in sports clubs]. Sportwissenschaft 44(1): 10–24.

30.

Frey

(1997) Not Just for the Money: An Economic Theory of Personal Motivation. Cheltenham, UK: Edward Elgar.

31.

Hallmann

(2015) Modelling the decision to volunteer in organized sports. Sport Management Review 18(3): 448–463.

32.

Hämmerle

Rullang

Pierdzioch

. (2014) Die Sozialfigur des Ehrenamt-lichen im Roten Kreuz – Ergebnisse einer vergleichenden empirischen Untersuchung [The social character of the volunteers in the Red Cross – Results of a comparative empirical study]. Working Papers of the European Institute for Socioeconomic, Nr. 8. Saarbrücken. Available at: http://www.soziooekonomie.org/fileadmin/soziooekonomie.org/pdf/EIS_Workingpaper_8_2014.pdf

33.

Hardin

Koo

King

et al . (2007) Sport volunteer motivations and demographic influences at a nationwide tour event. International Journal of Sport Management 8(1): 80–94.

34.

Haski-Leventhal

Bargal

(2008) The volunteer stages and transitions model: Organizational socialization of volunteers. Human Relations 61(1): 67–102.

35.

Haski-Leventhal

Cnaan

(2009) Group processes and volunteering: Using groups to enhance volunteerism. Administration in Social Work 33(1): 61–80.

36.

Haski-Leventhal

Meijs

LCMP

Hustinx

(2009) The third-party model: Enhancing volunteering through governments, corporations and educational institutes. Journal of Social Policy 39(1): 139–158.

37.

Heckathorn

(1989) Collective action and the second-order free-rider problem. Rationality and Society 1(1): 78–100.

38.

Heinemann

(1995) Einführung in die Ökonomie des Sports [Introduction to Sport Economics]. Schorndorf, Germany: Hofmann.

39.

Heinemann

(2004) Sportorganisationen: Verstehen und Gestalten [Understanding and Designing Sport Organisations]. Schorndorf, Germany: Hofmann.

40.

Hoeber

(2010) Experiences of volunteering in sport: Views from Aboriginal individuals. Sport Management Review 13(4): 345–354.

41.

Horch

(1998) Self-destroying process of sports clubs in Germany. European Journal of Sport Management 5(1): 46–58.

42.

Hox

(2002) Multilevel Analysis: Techniques and Applications. Mahwah, NJ: Erlbaum.

43.

Hustinx

Meijs

LCPM

(2011) Re-embedding volunteering: In search of a new collective ground. Voluntary Sector Review 2(1): 5–21.

44.

Ivancevich

Matteson

(2002) Organizational Behavior and Management. New York, NY: McGraw-Hill.

45.

Kim

Chelladurai

Trail

(2007) A model of volunteer retention in youth sport. Journal of Sport Management 21(2): 151–171.

46.

Kim

James

Connaughton

(2013) Comparison of volunteer motivations indifferent youth sport organizations. European Sport Management Quarterly 3(3): 343–365.

47.

Kristiansen

Skirstad

Parent

et al . (2015) ‘We can do it’: Community, resistance, social solidarity, and long-term volunteering at a sport event. Sport Management Review 18(2): 256–267.

48.

Lamprecht

Fischer

Stamm

(2012) Die Schweizer Sportvereine – Strukturen, Leistungen, Herausforderungen [Swiss Sport Clubs: Structures, Achievements, Challenges]. Zürich, Switzerland: Seismo.

49.

Maas

CJM

Hox

(2004) Robustness issues in multilevel regression analysis. Statistica Neerlandica 58(2): 127–137.

50.

Musick

Wilson

(2008) Volunteers: A Social Profile. Bloomington, IN: Indiana University Press.

51.

Nagel

(2006) Sportvereine im Wandel. Akteurtheoretische Analysen zur Entwicklung von Sportvereinen [Sport Clubs in Changing Times: Actor Theoretical Analyses of Trends in Sport Clubs]. Schorndorf, Germany: Hofmann.

52.

Nagel

(2007) Akteurtheoretische Analyse der Sportvereinsentwicklung – Ein theoretisch- methodischer Bezugsrahmen [Actor theoretical analysis of the development of sports – A theoretical-methodological frame of reference]. Sportwissenschaft 37(2): 186–201.

53.

Nagel

Conzelmann

Gabler

(2004) Sportvereine Auslaufmodell oder Hoffnungsträger [Sports Clubs – White Hope or Discontinued Model]. Tübingen, Germany: Attempto.

54.

Nichols

James

(2008) One size does not fit all: Implications of sport club diversity for their effectiveness as a policy tool for government support. Managing Leisure 13(2): 104–114.

55.

Nichols

Padmore

Taylor

et al . (2012) The relationship between types of sports club and English government policy to grow participation. International Journal of Sport Policy and Politics 4(2): 187–200.

56.

Okun

Schultz

(2003) Age and motives for volunteering: Testing hypotheses derived from socio-emotional selectivity theory. Psychology and Aging 18(2): 231–239.

57.

Peters-Davis

Burnt

Braunschweig

(2001) Factors associated with volunteering behavior among community dwelling individuals. Activities, Adaptation & Aging 26(2): 29–44.

58.

Peugh

(2010) A practical guide to multilevel modeling. Journal of School Psychology 48(1): 85–112.

59.

Pitsch

(1999) Ideologische Einflüsse in der empirischen Sozialforschung im Sport: aufgezeigt am Beispiel der Untersuchung von Sportvereinen [Ideological Influences in Empirical Social Research in Sport: Demonstrated by the Example of the Study of Voluntray Sports Clubs].Cologne: Strauß.

60.

Pitsch

Emrich

(1997) „Krise des Ehrenamts?“ Eine Analyse alter Daten [“Crisis of volunteering?” An analyses of old data]. Sportwissenschaft 27(4): 391–408.

61.

Pitsch

Emrich

(1999) Vom Nutzen der Sinn- und der Stoffhuberei. Eine Entgegnung auf die Stellungnahme von Heinemann/Schubert zum Beitrag „Krise des Ehrenamts?“ [On the benefits of sense and the material collection. A reply to the opinion of Heinemann / Schubert to post “crisis of volunteering?”]. Sportwissenschaft 29(4): 229–233.

62.

Raudenbush

Bryk

(2002) Hierarchical Linear Models: Applications and Data Analysis Methods. Thousand Oaks, CA: SAGE.

63.

Rehberg

(2005) Altruistic individualists: Motivations for international volunteering among young adults in Switzerland. Voluntas 16(2): 109–122.

64.

Robinson

(1950) Ecological correlations and the behavior of individuals. American Sociological Review 15(3): 351–357.

65.

Rotolo

Wilson

(2012) State-level differences in volunteerism in the United States: Research based on demographic, institutional, and cultural macrolevel theories. Nonprofit and Voluntary Sector Quarterly 41(3): 452–473.

66.

Scheerder

Vos

(2010) Sportclubs in beeld. Basisrapporting over het Vlaamse Sportclub panel 2009. Leuven: University of Leuven/Research Unit of Social Kinesiology and Sport Management.

67.

Schimank

(2005) Der Vereinssport in der Organisationsgesellschaft: Organisationssoziolo-gische Perspektiven auf ein Spannungsverhältnis [Club sport within the organisational society: Sociological perspectives to a strained relationship]. In: Alkemeyer

Rigauer

Sobiech

(eds) Organisationsentwicklungen und DeInstitutionali-sierungsprozesse im Sport. Schorndorf, Germany: Hofmann, pp.21–44.

68.

Schlesinger

Nagel

(2011) ‘Freiwilliges Engagement im Sportverein ist Ehrensache!’ Ein Modell zur Analyse der Mitarbeitsentscheidungen in Sportvereinen [‘Voluntary commitment to the sports club is a matter of honour!’ A model for analyzing decisions to volunteer in sports clubs]. Sport and Society 8(1): 3–27.

69.

Schlesinger

Nagel

(2013) ‘Who will volunteer?’ Analysing individual and structural factors of volunteering in Swiss sports clubs. European Journal of Sport Science 13(6): 707–715.

70.

Schlesinger

Nagel

(2015) Does the context matter? Analysing individual and structural factors of member commitment in sports clubs. European Journal of Sport and Society 12(1): 55–79.

71.

Schlesinger

Egli

Nagel

(2013) “Continue or terminate?” Determinants of long-term volunteering in sports clubs. European Sport Management Quarterly 13(1): 32–53.

72.

Silverberg

Marshall

Ellis

(2001) Measuring job satisfaction of volunteers in public parks and recreation. Journal of Park and Recreation Administration 19(1): 79–92.

73.

Skirstad

Hanstad

(2013) Gender matters in sport event volunteering. Managing Leisure 18(4): 316–330.

74.

Snijders

(2003) Multilevel analysis. In: Lewis-Beck

Bryman

Liao

(eds) The Sage Encyclopedia of Social Science Research Methods. Thousand Oaks, CA: SAGE, pp.673–677.

75.

Snijders

Bosker

(1999) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. London: SAGE.

76.

Stirling

Kilpatrick

Orpin

(2011) A psychological contract perspective to the link between non-profit organizations’ management practices and volunteer sustainability. Human Resource Development International 14(3): 321–336.

77.

Strigas

Jackson

Jr (2003) Motivating volunteers to serve and succeed: Design and results of a pilot study that explores demographics and motivational factors in sport volunteerism. International Sports Journal 7(1): 111–123.

78.

Studer

Von Schnurbein

(2013) Organizational factors affecting volunteers: A literature review on volunteer coordination. Voluntas 24(2): 403–440.

79.

Thiel

Mayer

(2009) Characteristics of voluntary sports clubs management: A sociological perspective. European Sport Management Quarterly 9(1): 81–98.

80.

Tönnies

(1963) Community and Society. New York, NY: Harper & Row.

81.

Wicker

Breuer

(2013) Understanding the importance of organizational resources to explain organizational problems: Evidence from non-profit sport clubs in Germany. Voluntas 24(2): 461–484.

82.

Wicker

Hallmann

(2013) A multi-level framework for investigating the engagement of sport volunteers. European Sport Management Quarterly 13(2): 110–139.

83.

Wicker

Breuer

Lamprecht

et al . (2014) Does club size matter? An examination of economies of scale, economies of scope, and organizational problems. Journal of Sport Management 28(3): 266–280.