From Sport Policy to National Federation Sport Policy: An Integrative Literature Review and Conceptualisation Attempt

: The notion of sport policy is a relatively recent subject of study, but it has been frequently used over the last 20 years, particularly referring to governmental sport policies. However, less research has been conducted in the ﬁeld of national sport federations (NSFs), which are key non-governmental structures in governing, organising and sustaining the development of sports. The objective of this article is to propose a conceptualisation of the notion of sport policy adapted to NSFs. The proposed approach consists of two steps: ﬁrst, we conducted an extensive integrative literature review around the notion of sport policy. Second, based on the literature review, we developed a model that conceptualises what an NSF sport policy is. This conceptual model is intended to be a useful analytical tool to design, implement and assess the success of a sport policy for both researchers and practitioners working in the ﬁeld of NSFs


Introduction
Sport is a broad term that comprises different dimensions, such as sport for all, elite sport, professional sport, Olympic sport, youth sport and sport for the disabled.Within the sport academic literature, sport policy as a field of study has been increasingly investigated by sport researchers over the last 20 years [1][2][3][4][5][6].In 2005, Houlihan noted that "while other policy areas have been the subject of extensive analysis, sport policy "has remained on the margins" [7] (p.163).The emergence, in 2009, of a specialised international journal (International Journal of Sport Policy and Politics, IJSPP (Initially International Journal of Sport Policy, 'and Politics' was added in 2011 [8]) is an embodiment of the growing importance of research in this area.It aims to: "Provide an outlet for articles that are concerned with analysing the role of government and the state in shaping the development of sport and how it is experienced in the contemporary world; [...] it is argued that there is a need to stimulate critical reflection on the nature and impact of state intervention in sport."[9] (p. 1) These first lines of the editorial of the first issue of IJSPP are one of the starting points of our article: research on sport policies focuses primarily on governmental sport policies at the national level, exploring the political dimension of sport and then analysing the political process and the effects of these policies.Yet, sport policies at a national level are not restricted to the governmental sphere.Voluntary or not-for-profit organisations at the national level, such as national sport federations (NSFs) or national sport governing bodies (NGBs) (NSF and NGB are synonymous; one is favoured over the other, depending on the geographic context, e.g., NSF is favoured in France, while NGB is favoured in the UK.In the present paper, we tend to favour NSF, following, e.g., Scheerder et al. in their edited book about NSFs from an international perspective, while using NGB specifically when this was done by the authors cited [10]), also have their own sport policies.This suggests a need to be clear on what we talk about when referring to sport policy, i.e., the problem definition for it.Hoekman and Scheerder echoed the argument and pointed out that "the fundamental difficulty of articulating an adequate problem definition for sport policy" remains [11] (p. 105).Therefore, this paper attempts to address this issue by highlighting the difference between governmental sport policy and NSF sport policy and, thus, providing clarity to the problem definition of the latter.Through our contribution, we highlight a paradox around the notion of sport policy, abundantly used, arguably as a self-evident notion, but, ultimately, not clearly defined.Pielke Jr and Harris [12] (p.138) concurred with this lack of precision around the concept: "Sport management, sport policy, sport governance-whatever terminology is used to describe research focused on decision making in and about sport.In our view, the polysemic dimension of the concept of sport policy deserves to be framed within the framework of key structures, namely sport federations." These organisations are indeed at the heart of sport systems and central to sport development and sustainability [13].Moreover, while having their own (organisational) sport policies, they also contribute to the achievement of governmental sport policy objectives [14,15].These two key roles of NSFs highlight the importance of having a clear understanding of their sport policy.However, to our knowledge, there is no framework or definition, let alone conceptualisation, of sport policy adapted to NSF, which we thus call NSF sport policy.Yet, similar to the commercial strategy of a company, the sport policy of a federation is the "core business" of its activity and, without a clear understanding of it, it will lead to ambiguity and conflicts within the policy-implementation process.This is likely to jeopardise the success of the NSF sport policy, hence underlining the importance of better understanding it.
The aim of this paper is to address the gap identified in order to help researchers and practitioners to better understand this subject, both in a prospective (conception) and retrospective (analysis) logic.Thus, the present research asks the questions: What does the notion of an NSF sport policy represent?What are the issues for the NSF?To answer these questions, we conducted an integrative literature review of publications that have revolved around the notions of sport policy, sport organisations and NSF.The rationale for the scope of the review going beyond NSF sport policy is two-fold.First, there is a need to clearly define an NSF as opposed to other sport organisations, which also need to be covered for the purpose of better identifying the specificities of an NSF, its link with other sport organisations and potential common issues.Second, although our focus is not specifically on governmental sport policy, which is the most covered in the literature, we acknowledge the need to review the literature on governmental sport policy for at least two main reasons: the literature on governmental sport policy can inform the sport policy dimensions to be considered in the context of NSFs, based on the assumption that different (levels of) sport policies (governmental, NSFs, etc.) have similarities due to their common focus on sport; and the sport policy applied by an NSF is likely to depend on the governmental sport policy if the NSF needs governmental funding to sustain its existence and/or the NSF aims to obtain governmental funding to generate more revenue, meaning the NSF sport policy needs to align (at least to some extent) with the objectives set by the government/in the governmental sport policy.
The main objective is to review these elements and see to what extent they can be adapted to sport federations.More specifically, the research objectives are:

•
to review the literature on governmental sport policy to derive some findings relevant to NSF sport policy; • to review the literature on sport policy in the context of sport organisations other than NSFs to identify some issues relevant to NSF sport policy; • to review the literature on NSFs, both on their specificities, to clearly define this type of sport organisation, and also more specifically on their sport policy; • to integrate the knowledge from the first three research objectives to derive a conceptual model to better understand an NSF sport policy.
The following section presents the methodology of the literature review process.Afterwards, we depict the notion of sport policy, highlighting the main characteristics of a sport policy, while evidencing a conceptual gap for the NSF sport policy.We then propose a conceptual model to better understand an NSF sport policy, before a detailed discussion of this model in its capacity to be a useful tool for both researchers and practitioners.

Methodology
Our methodological approach is threefold.First, we conducted an integrative literature review [16][17][18] by searching for all significant contributions on sport policy.In line with Whitemore and Knafl, a five-step approach was applied: problem identification (presented in the introduction, i.e., how to better define sport policy and, more specifically, an NSF sport policy?), literature search, data evaluation, data analysis and presentation of findings [18].

Literature Search
We conducted a literature search of key research-specialised databases using keyword search terms, both in English and French.The keywords in English (not provided here, but equivalent in French) were: "sport policy(ies)", "sport politics", "politics and sport", "policy-making and sport", "sport federation(s)", "sport governing body(ies)", "national sport organisation", and "sport development".The search was conducted from April to October 2022.The objective was to identify the literature providing elements contributing to the understanding, analysis, description (exploration) and even conceptualisation around sport policy.This was to review the issues and the boundaries and to be able to better understand them for a national sport federation.In other words, as long as the references included elements that allowed for a better understanding of the issues of any sport policy, and more generally, of the development of sport from a local to international dimension, they were taken into account.
Second, before conducting the data evaluation, data analysis and presentation of findings from the literature, we worked by snowballing, looking at the references of key articles that emerged from our previous approach via search engines (backward snowballing), as well as their citing publications (forward snowballing) [19][20][21].We do not claim to have listed all the articles on sport policy, but these are significant references that can help answer our questions.No boundaries were defined for publication dates.
Third, after conducting the data evaluation, data analysis and presentation of findings from the literature, we relied on the latter findings in our attempt to conceptualise an NSF sport policy.

Data Evaluation
The evaluation process consisted of two steps, consistent with the snowballing approach explained above.The first step consisted in collecting references that dealt directly with sport policy.It should be noted that a simple search on Google Scholar with the terms "sport policy" brings up 348,000 results; by activating the "sort by relevance" option, we obtained 792 references.Then, we selected only the references that included at least one chosen keyword in the title (e.g., "sport policy").Similar strategies were applied to other search engines.Published work written in English or French were included in the review, while references such as opinion pieces or newspaper articles were excluded.Then we conducted a step-by-step review by reading carefully information contained in the abstract first, then by acknowledging the publication outlet and finally making an in-depth review.This first step brought out 56 articles, 12 books, 4 book chapters and 1 editorial article, which served as a basis for the next step.
The second step focused on identifying significant books or articles that we would have missed during the first step.These do not fit in the first selection condition (title with keywords side by side), but fit perfectly in the objective of the literature review.Thanks to the backward snowballing approach, we were able to identify other references.For example, the work of Coalter [22] is useful, but its title does not have one of our keywords.This work was identified via the references of Keat and Sam [23] or Houlihan et al. [9].As long as these new references were significant for our research aim, we integrated them into our literature review.This second step resulted in 29 articles, 5 books, 1 book chapter and 1 book review.
The total of 109 references (of which 7 are in French, namely, 4 articles and 3 books) obtained are, therefore, within the scope of our keywords and/or the objectives of this review.It is acknowledged that this specific scope represents one of the limitations of this literature review, as authors producing a review on the same topic may identify different references with another scope.However, as noted by Torraco [16] and consistent with Schulenkorf et al., "an integrative review does not presuppose to be exhaustive on any given topic or topics, but rather exhaustive within its predetermined and declared boundaries."[24] (p.24).

Data Analysis
The process of data analysis was conducted through the review of the full text of each reference.The data analysis process for the 109 publications continued, following Miles and Huberman's recommendations, which suggested dividing data into intellectual bins that ultimately reflect common themes between books and articles [25].Under the overarching subject 'sport policy', two streams were distinguished: governmental sport policy and sport policy from non-governmental national sport organisations.Then, each stream was analysed and categorised into four levels.The first stream (governmental sport policy) was split into two fields of study, i.e., development of and through sport, while the second stream (non-governmental sport policy) was split into two ways of exploring sport policy, i.e., in the context of a national sport federation and outside (e.g., international federation, national Olympic committee and local clubs).Both streams were then split into three different approaches, i.e., conceptual (attempting to conceptualise what sport policy is), descriptive (describing the sport policy in a country or a sport organisation) and analytical (attempting to understand the relationships between actors, processes and/or effects of a sport policy).The third level of analysis was the research topic covered in the publications analysed.Last, the fourth level was about the issues to be considered as identified in the research reviewed.
Table 1 summarises this data analysis process.There are common issues to be considered between the two fields of study within each stream, explaining why these issues are grouped together at the stream level for the fourth level of analysis rather than distinguished between fields of study, as done for the second and third levels of analysis.This four-step analysis process has allowed us to comprehend how the topic of sport policy has been treated in the main international journals and books.Table 1 highlights the fact that no conceptual approach to sport policy has been made in any context other than public/governmental.Based on this gap, we intend to highlight elements of conceptualisation, while underlining the issues to be considered during this conceptualisation.
It is important to clarify that while every attempt was made to reach an appropriate categorisation, we acknowledge that our manual classification process remains imperfect and, therefore, represents a limitation of this study.It should also be emphasised that, for each level of analysis, a reference was only counted for one category.Yet, several articles that are about governmental sport policy also describe the overall sport system of a country and, thus, deal with some issues about non-governmental sport policy.In this case, we have chosen to categorize the reference in "governmental sport policy".Moreover, when other authors focus on an NSF sport policy and emphasize the link between national governmental policy and an NSF sport policy, we have categorized it under the "nongovernmental sport policy" stream.

Conceptual Foundations: Findings from the Literature
The presentation of our conceptual foundations follows our analysis, i.e., we first review the research on governmental sport policies, then on sport organisations policies.

Sport Policy as Public Policy: A Polysemous and Multidimensional Subject
As a public policy, the stakes of a sport policy go beyond the development of sport.Thus, we will show that the motivations are both domestic and diplomatic.Then, we will highlight the predominance of descriptive and analytical approaches regarding governmental sport policies.

From Sport Policy to Development of and through Sport
Deconstructing our subject of study is enough to understand that it covers two polysemous notions (sport and policy).Therefore, when we focus on the notion of sport policy, it is hardly surprising to see the multitude of associated terms.For example, if we refer to Green [26,27], it is interesting to see that the notion is directly associated with 'statements', 'priorities', 'sector', 'commitments', 'interventions', 'discourse', 'concerns' and 'processes'.Other terms are a testimony to the richness, but also the complexity of the concept we are dealing with.Indeed, more broadly, the concept is directly associated with the notions of objectives [8,28], strategy [29], initiatives or model [30], decisions [31], factors [32,33] and also delivery [34].
Green and Collins [31] build on Houlihan [1] (1997) to argue that "in many countries sport and government have been inextricably linked across a diverse range of policy issues including health social inclusion, community development, education and the achievement of elite sporting success" [31] (p.225).It was around this time that sport was considered a public policy area.Reiche admits that this sector constitutes a good platform for transmitting messages to the public [35].This consideration is in line with Hogwood and Gunn, who suggest that "for a policy to be regarded as a 'public policy' it must to some degree have been generated or at least processed within the framework of governmental procedures, influences and organisations" [36] (p.24).From then on, Houlihan proposes models for the analysis of sport public policies, in order to account for the effect of sport and, ultimately, of the sport policies put in place [7].The author also regrets that too little literature "utilized the extensive array of concepts, analytical frameworks and theories developed in mainstream policy analysis to aid understanding of sport policy-making and the role of government" [7] (p.164).
As a result, public policy approaches have multiplied.In this respect, several authors invite us to make the connection between sport policy and public policy, in order to identify similar characteristics between them [7,[37][38][39].Thus, we have been able to find several resources that allow us to better account for both the structure and content of sport policies.
For example, Callède identifies five characteristics that describe what a public policy is, and then transposed them to a sport policy [40].The latter would then be the display of (1) a general framework for action, (2) goals and objectives to be achieved, (3) concrete measures, (4) audiences to be reached and, finally, (5) the allocation of resources and/or regulatory prescriptions including coercive modalities.Later, Barget and Vailleau define a policy as "a set of decisions and rules of conduct adopted in advance, for a certain period of time, with a view to achieving certain general objectives" [37] (p.21).More recently, Bayeux suggests a categorisation into four levels to understand a sport policy: (1) political (definition of values and finalities), ( 2) strategic (set of coordinated actions related to objectives aligned with the finalities), ( 3) tactical (allocation of resources, management and organisational modalities) and ( 4) operational (concrete implementation of actions) [41].
Girginov also considers sport policy as a "field of public policy" and defines it as follows: "Sport policy will be interpreted here as a framework of principles, objectives, and planned (or unplanned) actions (or inactions) developed by credible local, national, or international agencies aiming to achieve common ends concerning specific communities within a given range of resources, time, and space."[42] (p.176) The basis of these definitions, as well as the multitude of associated notions that we have seen above, is thus the link to objectives of the development of sport (or "sport development"; [43]) and through sport [2]).This is consistent with Shilbury et al., who note that the concept of sport policy overlaps with these two other fields.According to the authors, these three fields share the common feature of motivating individuals to participate in sport [3].Consequently, it seems necessary to state the content that governs sport policies.In this regard, Houlihan and Zheng summarise the political functions of sport into two motivations, domestic and diplomatic [44].These two aspects legitimise the implementation of sport policies [45].We elaborate on these two functions below.

Domestic vs. Diplomatic Motivations
At the level of domestic impacts, we can synthesise them into two dimensions.On the one hand, there is the duality of sport for all and elite sport, highlighted by Green [26,27] and Sam [46], as 'dominant ideas' in sport policy (intrinsic value of sport).On the other hand, there is the dimension of the social-economic and societal effects of sport on a given territory (instrumental value of sport).
Many authors have analysed the intrinsic value of sport in various contexts.The "quantitative" development of the number of sports is often the basis of sport policies.Nicholson et al. refer to "sport participation" [47], while Skille refers to "sport for all" [30].Zheng et al. indicate that this is a real issue for the Chinese case, with sport for all becoming more prominent in recent years; "it has received increased government attention in recent years" [48] (p.484).Shilbury et al., distinguish between studies on mass participation versus elite sport, while also emphasising studies that specifically question which sport policies contribute to the success of elite sports, and how they manage to last on the international scene [3].On the topic of mass participation versus elite sport, Sam [46] questions the complementarity of these two dimensions and highlights the contradictory nature of policies that simultaneously promote gender equality and international success.
Jedlicka et al., explore how political regime type influences elite sport policy development and, thus, look at the state intervention in sport and its consequences in terms of elite success [49].On the promotion of elite sport, we also cite Green and Houlihan [50], Green and Oakley [51] (2001), De Bosscher et al. [32,52], Andersen and Ronglan [53], and Dowling et al. [54] as relevant references in this area.For example, De Bosscher et al. [32,52] identify the policy factors that determine elite sporting success.The search for and promotion of elite sport also includes the sport talent identification process [55] and, thus, the investment in young talent.
Concerning the social-economic and societal effects of sport in a given territory, the literature highlights the extent to which sport allows the achievement of non-sporting objectives: healthy lifestyle, fight against obesity and sedentary lifestyle, regional planning and social cohesion.In other words, sport contributes to wide-ranging development outcomes, which are cross-cutting issues [56].Hoekman et al., state that "sport is positioned more and more as an instrument with health and social effectivity to help achieve non-sport policy goals" [11] (p. 89).Jaccard et al. [57] do not forget the educational, cultural, tourism and sustainable development issues associated with a sport policy, using the case of a municipality as an example.
This instrumental value of sport is also found in external motivations, which can be described as diplomatic.Houlihan talks about 'politics and sport' to refer to investments, regulations and interventions in the sport sector by governments [58].He differentiates this from "politics in sport", which analyses the role of sport governing bodies that resolve issues that arise within sports themselves.As an effective political tool [12,59], sport issues have gained higher priority on government agendas.Thus, some references analyse how sport has come to be used as a political means to achieve sport-external ends (see e.g., [60]).As a result, sport is increasingly seen as a diplomatic « soft power » resource [12,[61][62][63][64][65][66][67].Houlihan and Zheng demonstrate how sport can be used for image building and building of tentative diplomatic links, and even express support or displeasure for the actions of other states [44].

Descriptive vs. Analytical Approaches
In parallel to these conceptual works, which have made it possible to characterise public sport policies, others have drawn on the concept, in a descriptive logic on the one hand, and an analytical one on the other.Descriptive articles have emerged, notably in IJSPP, with articles describing the sport policy of a given country.All continents are represented: Europe [68][69][70][71][72][73][74][75], Asia [48,[76][77][78], America [38,[79][80][81], Africa [82-85] and Oceania [86].The structure of these papers is similar: presentation of the country's political system and/or organisation, the place of the sport system, the national sport organisation, the distribution of resources, and then the evolution over time of the policies carried out in the field of sport and the priorities identified.Nevertheless, some particularities should be highlighted, namely, the religious (or even denominational) dimension [77], cultural dimension [71] or legislative evolutions [38].In a similar vein, Bergsgrad et al., propose a comparative analysis of sport systems, focusing in particular on the processes of policy change and issues associated with policy convergence [87].Later, Dowling et al. analyse the stakes of the comparison between sport policies [88] (following Hallmann and Petry [89]) from philosophical and methodological perspectives.They highlight the dominance of positivist approaches, which put the emphasis on causality and outcomes when comparing state sport policies.They regret that researchers do not take enough account of the social, economic and political contexts of the different countries that are compared.
In summary, the common point of these contributions is a focus on the state dimension and not an approach of non-governmental organisations, and even less federations/NGBs.Thus, there are attempts to define a sport policy, but restricted to the governmental dimension, with suggestions to apply it to non-governmental organisations.Although the definitions of governmental sport policies may have some relevance to non-governmental organisations, it is paramount to properly characterise these organisations and understand their stakes.Table 2 summarises the findings presented above about governmental sport policy.Processes of policy change and issues associated with policy convergence Dominance of positivist approaches which put the emphasis on causality and outcomes Lack of consideration of the social, economic and political contexts of the different countries Source: authors.

Non-Governmental Sport Policy: Issues to Be Considered
This section is structured in two parts.Literature on sport organisations other than NSFs is presented first, followed by publications that have focused specifically on NSFs.

Non-Governmental Sport Organisations Other than NSFs
When looking at the sport policy of non-governmental sport organisations, Bayle distinguishes organisations whose purpose is to federate a number of organisations and to provide them with services (e.g., (inter)national sport federation, (inter)national Olympic committee, professional sport league) and organisations that are only local ones (e.g., club, amateur or professional) [90].
Whatever the typology of the non-governmental sport organisations studied, it appears that their sport policy is sectoral and transversal.Jaccard et al. [57] draw on Bayeux's characterisation to support this dual characteristic: sectoral (development of sport by the sport organisation in question), but also cross-cutting, insofar as it interferes closely with other issues [41].This is because their role is multidimensional: identifying talent, supporting performance development and delivering competitive success, but also foster regeneration, better health, more cohesive and inclusive communities, and lifelong learning [41].
In this section, we present the different issues expressed around sport policies related to the Olympic and Paralympic movement, sport organisations (SOs) other than NSFs and the sport clubs.Moreover, whether it is in the Olympic and Paralympic movement at the international level or in local clubs, the references studied highlight governance issues, which impact the decision-making process.
Olympic and Paralympic Movement Policy: How to Position Disability and Traditional Sports, Issues to Be Considered by NSF? Hums et al. address the governance issues of disability sport, initially represented by Disabled Sport Organisations (DSOs), and, in particular, the legal and policy implications for integration of athletes with disabilities into the United States Olympic Committee (USOC) and NGBs [91].Regarding their question "What are the ways the USOC, the NGBs, and the DSOs proposed to integrate sport for people with disabilities into the USOC and NGBs?" [91] (p.266), the authors describe, in particular, the creation in 2000 of the United States Paralympic Sports Corporation (USPSC), which happened without a concrete plan for the implementation of such a Paralympic organisation.DSOs were unsure what roles and responsibilities they were going to have.These elements have led to governance and leadership crises, with effects on the relationship between organisations and, thus, the political and decision-making process.Han et al. explain the difficulties for the International Wushu Federation (IWUF) to bring its practice into the Olympic environment [92].Beyond the elements of lobbying and international governance, the authors' analysis shows the ambiguity of the IWUF's Olympic policy over the years, insofar as this discipline is marked by a strong tradition, particularly in China, whereas entry into Olympism suggests a stronger opening to the world and a certain modernity.
Patatas et al. propose an overview of the sport policy factors (e.g., financial support, organisation and structure of Parasport) that could influence the development of paraathletes' career pathways in Paralympic sport, while mentioning the impact of stakeholders on the policy development and implementation: "policymakers and stakeholders involved in policy processes need to critically position disability within policy decision making to deliver successful para-athletes' pathways" [33] (p.949).Sport Policy of SOs Other than NSFs: Some Considerations Relevant to NSF Houlihan and Green examine the modernisation of UK Sport and Sport England, indicating that the transition to a new sport system has resulted in a narrowing of these two organizations' objectives, the adoption of business-like principles and a 'command and control' regime in relationships with their sport policy stakeholders [29].Strittmatter et al. argue the idea that the sport policy process can be understood as an interlinked chain of legitimating acts, i.e., legitimacy and legitimating acts as key to all phases of the sport policy process [93].Legitimacy is fundamental to the decision-making process, as it implies the ability of actors "to engage in proper conduct and to be a model for others' conduct" [93] (p.623).In this way, legitimating acts connect the last phase of the policy process with the first in a cyclical manner.Green and Houlihan explore the process of elite sport policy change in two sports and their respective NSOs and NGBs, making a comparison between Canada and the UK [50].They argue in particular that there is stronger evidence of an elite sport advocacy coalition in the UK than in Canada.Myers and Doherty [94] (p.326) indicate that "policy is intended to address issues that relate to public concerns".Their contribution characterises NSO strategies and initiatives that represent individual, structural and cultural perspectives regarding the Sport Canada Policy on Women in Sport.It highlights that decision making in terms of orientations is contextualised and addresses a population that is well-identified beforehand.Myers and Doherty also refer to the elements of form and the need for a document that is intelligible and understandable to all: "The language of the policy must be clear enough to identify specific initiatives and intended outcomes yet, at the same time, be broad enough to allow those same individuals and organisations to create initiatives to address their unique situations" [94] (p.340).
These elements make sense with the recent contributions of Han et al., who analyse the International Wushu Federation sport policy with an analytical logic model [92] developed by Chen et al., which provides a specific framework for policy analysis and evaluation [95].It incorporates the following elements: 'context/environment' (e.g., social or political context); 'inputs/resources' (e.g., financial and human resources); 'activities (throughputs)' (i.e., kinds of actions); 'outputs' (e.g., numbers of participants); temporal horizon (short-term and long-term); 'outcomes' (e.g., changes in behaviour); and 'impact' (i.e., broader changes in organisations, communities or societies).According to Strittmatter et al., who look at the national Olympic movement (Norwegian confederation of sports) and international bodies (World Skate and Fédération Internationale de Football Association), the sport policy process is iterative, with four phases: (1) problem definition and agenda setting, (2) policy formulation and (3) policy implementation, then (4) policy evaluation [93].

Sport Clubs: Inconsistent Implementation Objectives Due to a Lack of Clarity of NSF Sport Policies?
The path from national development policy to local implementation requires clubs to interpret or even modify central policy to fit the local context [96].The contribution of Harris et al. highlights the key role played by voluntary sports clubs in delivering both government and NGBs sport policies and the pressure they face [56].Nagel et al. (2016) look at the situation of sport clubs in Europe [97].They indicate differences in approaches in terms of social movements that form the basis of sport clubs, while admitting that a sport club, whatever the national model of sport, tends to mix three approaches: the concept of competition and comparison of achievement; the idea of promoting body, health and mind; and training the human body.Despite being central to the delivery of these policies, Harris et al. suggest that voluntary sport clubs are not clear about the policy objectives for sport [56].Thus, they recommend that top-down policymakers develop a targeted approach, requiring research and a deeper understanding and respect for the diversity of club types.
In line with Skille, it is worth noting that club sport policies have mainly been apprehended within the framework of NSFs, whose main contributions and issues we present hereafter [30,96,98].

National Sport Federations: From Their Specific Environment to Implementation Issues
The emergence of NSFs coincided with the evolution of local sports clubs [99].The growing popularity of sports led to the need to create "umbrella organisations" to govern sport, supervise its practice, ensure its promotion and development at all levels, and guarantee the organisation of competitions [14].Generally speaking, the role of NSFs is to ensure that the rules of the game in the sports they promote within their country are respected, and to represent them, as well as those who participate in them.
Studying any NSF means first considering the specificities of their external and internal environment.At the external level, an NSF has three institutional partners with different logics, objectives and levels of impact: the international federation, the National Olympic Committee and the government, represented, for example, by a ministry of sport [100,101].
At the internal level, the particularity is the very different actors involved.Viollet et al., distinguish between sport technicians, administrative staff and politicians (the vast majority of whom are volunteers) [102].Moreover, in terms of structure and unit of analysis, it seems necessary, as Bayle mentions, not to limit an NSF to the headquarters, but to consider it as a system (or network) [90].At the national level, Bayle highlights the logic of collaboration between an NSF headquarters and local components when dealing with sport policy [90].As a national association of regional and local associations, an NSF appears to be a really organised community.To manage their network of affiliated clubs, federations use relays.These three components (federations, relays and clubs) form an NSF network.The analysis of the work on sport policies highlights the rise of NSFs in the sport sector, with consequences in terms of professionalisation and managerial culture.Thus, the key issues for NSFs lie in the close link they have with the national (public) sport policy, and the need to involve clubs in all aspects of their development.

From Voluntary Structures to Professionalisation and Subsequent Performance and Managerial Cultures
Harris et al., emphasise the particularity of NSFs as voluntary organisations: the voluntary membership, the realisation of the common interests of the members, the democratic decision-making structure, the possible mix between paid jobs and voluntary work, the autonomy and the not-for-profit orientation [56].Kikulis develops a description of three aspects of governance and decision making (the institutionalisation of volunteer boards, the deinstitutionalisation of volunteer control and the semi-institutionalisation of paid executive roles) that provides an alternative institutional explanation of continuity and change in governance and decision making in NSOs [103].Regarding this issue, Harris et al., point out that, at the national level, the NSFs are not primarily companies driven by economic criteria, but that these structures have become more professionalised in response to institutional pressures [56].As key structures in the development of their sport, NSFs, "need to establish contemporary management structures and programmes to accomplish their work more efficiently and to adequately meet the expectations of a complex and dynamically changing environment".NSFs have, thus, not been exempted from the phenomenon of professionalisation [104] (p.408).Nagel et al., break down the professionalisation of NSFs into several fields: their activities, individuals, structures and processes [104].They define the professionalisation of sport organisations as an organisational process of transformation leading towards organisational rationalisation, efficiency and business-like management.
This logic of professionalisation is part of the dynamic of 'modernisation' seen above, notably in the article by Houlihan and Green [29].This involves clarifying objectives and relating structures and roles more closely to them (usually involving specialisation, bureaucratisation and the introduction of more paid professionals).This professionalisation of NSFs is not limited to the rise of paid individuals, but also the rise in competence of internal actors.Bayle, in his study on the governance of French NSFs, insists on the need for skilled human resources for these structures, as the media and economic stakes have become substantial for some of them [100].Bernardeau Moreau also studies the roles of the leaders and the impact of the NSF professionalisation on these roles [105].This professionalisation in governance is reflected in the emergence of new essential questions concerning the 'quality' of leaders, the separation of powers between politics and management, the composition and functioning of boards of directors, the relationship between (elected) directors and managers, and the capacity and willingness to steer the performance of these organisations.Bayle stresses the importance of performance management in NSFs, which reflects real specificities [90].Indeed, NSFs moved from a management of "organisational culture" (management by values) to a management of "professionalization", i.e., of the relationship between volunteers and paid professionals.
With this rise in power of NSFs, the challenges lie in managing the coexistence of several actors with different profiles.For example, Nam et al., explain the conflicts among stakeholders (administrators, coaches and student-athletes) regarding the new academic system in the Korea University Sport Federation [106].In the context studied, they show that conflicts are often produced due to the power relations between the ruling class and the ruled class.Skille and Chroni focus on the Norwegian sports federations' organizational culture [107].According to the authors, the organisational culture has different dimensions: the culture of the sport itself, the organisation's overall success, the members of the organisation and the organisation's goals.They show that the organisational culture impacts the success of a country in terms of sport performance because it influences the work of the coaches.
Girginov examines the formation of a new Bulgarian sport policy, based on the cases of the Aerobic Union and the School Sports Federation [42].He identifies the key sport policy actors and the strategies and strategic relations they are pursuing in shaping sport policy.Girginov concludes his analysis by stating that state-sport relations were "responsible for shaping the communication process in policy making in a "top-down" manner, thus nurturing further the conceptual and structural selectivity of the sport system" [42] (p.193).The author also refers to the notion of "strategic selectivity" to describe Bulgarian sport policy, which has favoured elite sport's national and international developments and largely neglected policy making and the 'sport for all' approach at the local level.Girginov shows the extent to which NSFs can be impacted by the national (public) policy in sport systems where the government holds a prominent position due to its political and economic weight [42].

Link between National (Public) and NSF Sport Policy
Girginov [42] (p.173) states: "sports policy is a strategic relation, the formation of which needs to be viewed in a broader context of state-society relations".Several authors have highlighted the link between national sport policy (state, government and ministry of sports) and NSF sport policy.In order to analyse the success of an NSF sport policy, Bayle and Durand focus on the relationship with the state, and so its public sport policy.According to them [101], NSFs have difficulties in maintaining coherence between professional sport and high-performance (elite) sport, even though these two elements are often the pillars of national policies.They also explain that the relationship between the state and NSFs is variable, depending on the economic weight and the level of professionalisation of the NSFs.
The NSFs that are doing best on these two dimensions have become more and more autonomous and developed partnerships with companies, which can force them to take on new expectations.Bayle and Durand mention, for example, new objectives, such as regional planning [101].
The comparative analysis of the book chapters by Bayle about Switzerland [14] and Scelles [15] about France highlights the difference between countries.The Swiss case is very specific, as the NSFs' goals are defined by the Swiss Olympics in partnership with the NSFs rather than by the government.By contrast, the latter strongly influences NSFs' goals in France and other countries [15].Dowling et al., show how, in Canada, many NSFs rely heavily upon government funding [108].Regarding Athletics Canada, the authors note that: "it is evident that many of the developments that have occurred within Athletics Canada over the past decade have been governmental-driven" (p.95).In China, Peng et al., (2019Peng et al., ( , 2021) ) described the Chinese Football Association as a pioneering NSF reforming to become an autonomous national governing body insulated from the long-lasting government influence [109,110].
Viollet et al., show the extent to which the involvement of actors in the design of an NSF's sport policy is decisive [102].The leadership of the actors representing the state in the federations allows them to take the reins of the project and, thus, aligns it with governmental expectations, even though the French Rugby Union (the case studied by Viollet et al.,) is not financially dependent on the French Ministry of Sport [15].However, multiple leadership (e.g., government vs.NSF) can cause managerial confusion for implementers, as underlined by Peng et al., (2022) in the case of Chinese youth football [111].
In the context of economic hardship, Giannoulakis et al., examine implications of austerity measures and reduced public spending on elite sport development through the case of NSFs in Greece [112].They demonstrate that gradual decrease in government funding for elite sports in Greece decline both elite sport performance and the sport talent identification process alike.Giannoulakis et al., also explain that this economic context has an impact on clubs, which must "generate additional income in order to address their operational costs and become less dependent on public funding" [112] (p.94).

NSF Policies: Implementation by Clubs
The path from national policy to local implementation is 'long and difficult' [96] (p.181), requiring implementing structures to interpret and modify centrally developed policy to suit the local context.Clubs play a key role in this implementation logic.Among his three theoretical approaches to studying sport policy, Skille proposes a top/down implementation model, where the starting point is the definition of objectives by policymakers and the allocation of resources [96].Then, a set of three variables come into play: (1) characteristics of the implementers, (2) organisational communication and (3) economic, political and socio-cultural conditions.Then, the implementation depends on the willingness of the implementers, i.e., their ability and willingness to achieve the intended result in relation to the set objectives.Skille's analyses show that sport club volunteers have an absolute loyalty to their local environment and interpret centrally developed policy through the filters of the local context, organisational characteristics and personal experiences [96,98].
It is important to take into account that the situation and environment of clubs is variable and dynamic, both within and around the club, whereas sport policy is a generally non-variable process (at least, less dynamic than local relays).These heterogeneous realities of clubs contrast with the homogeneous conditions that policy assumes [28].
Thus, the challenge for political actors is how to integrate the diverse meanings and interpretations of policies with the purpose they had when the policy was born.To do this, the involvement of intermediary structures is a facilitating factor.Following their analysis of the German Swimming Federation, Fahrner and Klenk emphasise the collaboration of members in the whole network of an NSF.According to them, not only local sport clubs, but also regional governing bodies, play a crucial role as intermediaries in the implementation context, bearing responsibility for (un-)successful multilevel policy implementation [113].Another interesting element of the Fahrner and Klenk study is that the inability of an NSF to give binding directives at all organisational levels is perceived only as pressure, which can cause conflicts and distrust [113].Relying on Skille [96], Viollet et al., in their study of the French Rugby Union argue that different groups of actors act at different levels: "some of them are involved in the conception of the sport policy, while others organize the conditions to achieve the objectives while others take responsibility for implementing these objectives through actions 'on the field'" [114] (p.321).
In this regard, the strategic role of the national actors is impacted by their interorganisational relationships [102].Hence, it seems necessary to generate a more collaborative partnership with the regional associations by introducing a power-sharing approach.It is also important to note that political actors need to give implementers sufficient time to make sense of the issues and to gauge their implications for the context, and to build capacity to respond locally in an appropriate manner [28].Therefore, the implementation in the club of the central policy can be seen as the meeting of 'top-down' initiatives and 'bottom-up' demands [115].Different constructions of meaning among local 'implementers' can generate policy outcomes quite different from the original intentions of policy actors, as implementers incorporate mainstream policy into local contexts.
Table 3 summarises the main topics and references regarding studies about nongovernmental sport policies.A processual (from problem definition to policy evaluation) and analytical approach (to enable evaluation)

•
The conditions for implementing the policy must be considered at the time of its definition Source: authors.
As this section confirms, research on sport policies is rich in information and takes place in various sport contexts [117].Several elements help us to better understand a sport organisation's sport policy, particularly with regard to decision making and analysis processes.Our state-of-the-art review brings out indications on the "life" of a sport organisation's sport policy.Further, when we look at the national level, we can see that governance issues are very important, as these organisations must deal both with international structures and local clubs.
In relation to NSFs, the notion of sport policy is addressed, without being clearly defined.Thus, we have tried to characterise these organisations and their evolution.With regard to their sport policy, it appears that they must reconcile the interests of stakeholders at the design stage (notably the government/national policy vs. local realities) and make their structures (regional, departmental and local) allies, so that the sport policy is implemented throughout the national territory.
In the end, several issues emerge when we look at a sport policy in the context of a sport organisation.The first type of issue is related to power.As soon as a strategic process of reflection is established, questions of leadership and legitimacy between the decision-makers emerge.Regarding NSFs, this raises questions about their governance and the way in which they involve different actors in large-scale projects.This governance issue is linked to a second issue about the relationship with the state and the impact of public resources (human and financial) allocated to NSFs.Moreover, like public sport policies, sport organisations are committed to considering both non-sporting and sporting objectives, the latter having to reconcile both performance and sport for all.
Other issues to be raised are methodological.Like any political process, there is a processual approach and different levels of analysis, all within a logic of continuous learning (i.e., being able to evaluate the sport policy to improve its effectiveness over the years).Finally, the conditions for implementing the policy must be considered at the time of its definition, which requires a detailed knowledge of local conditions.
To summarise, our review of the literature has highlighted issues of space (actors, NSF network, governance and national and local dimensions), content and methodology (process and analysis).This highlights the need for a of analysis instead of partial definitions or characteristics concerning NSFs.Our model follows seven overarching elements identified as key in the literature review: • A multi-level approach: depending on the configuration of the NSFs, the structures making up the 'federation network' must be considered, i.e., headquarters, decentralised structures and clubs.Caldwell and Mays use a macro-meso-micro levels approach to study a policy implementation [118].In regard to sport policy, Dowling et al. also consider this approach to distinguish units of analysis [88].As such, in the NSF context, our proposal is to consider, respectively, the headquarters, the intermediate/decentralised structures and clubs at the macro, meso and micro levels.

•
A suggested definition of an NSF sport policy (inspired by the literature on governmental sport policy, while being adapted to an NSF) as a set of principles, intentions and strategic orientations aiming at developing quantitatively and/or qualitatively the (main) sport (e.g., football) and sometimes associated disciplines (e.g., futsal).

•
These strategic elements need to be translated into objectives, a specification of the means allocated to achieve the objectives (distribution of resources) and an action plan.

•
Concerning the operationalisation of the NSF sport policy, a distinction must be made between actions inherent to elite performance, which are led by the NSF headquarters (coaching of national teams, high-level athletes and preparation for international competitions), and other actions relating to the development of and through sport, for which all structures can intervene.Seippel and Belbo underline how clubs are key structures in the development of sport for all and the achievement of non-sporting goals through sport [119].

•
Like any policy, there must be a "transformative" aim, echoing the evolutions of sport policy and processes of policy change covered in the literature on governmental sport policy.A sport policy has a driving and/or dysfunction-reducing dimension, i.e., an impact on the internal and external environment.These elements refer to the 'outputs', 'outcomes' and 'impact' dimensions developed by Chen et al. [95] and Han et al. [92].

•
A sport policy is a multi-stage process.We can build on Strittmatter et al. to characterise four main phases: (1) problem definition and agenda setting, (2) policy formulation, (3) policy implementation and (4) policy evaluation [93].This classical view of the process can be further developed (see, for example, Bridgman and Davis, [120]).For phase 1, a diagnostic phase can be added, i.e., an analysis of the current policy.Phase 2 can include consultation with stakeholders, coordination of resources, approval of the various components and consistency with the social, economic and political context of the country [88].These elements can also be included in phase 3.In phase 4, the revision of the sport policy can be added, notably in the logic of iterativity expressed by Strittmatter et al. [93].In particular, its ability to adapt to local realities can be evaluated and further improved.
• Finally, Myers and Doherty refer to the elements of form and the need for a document that is intelligible and understandable to all [94].In this vein, Piggin proposes a series of questions to analyse the political discourse applicable to sport policy, both in terms of policy production and dissemination [121].We can suggest that the aesthetics of the document produced may also play a significant role, e.g., in helping give meaning to actions.
Figure 1 summarises our conceptualisation attempt.
iterativity expressed by Strittmatter et al. [93].In particular, its ability to adapt to local realities can be evaluated and further improved.

•
Finally, Myers and Doherty refer to the elements of form and the need for a document that is intelligible and understandable to all [94].In this vein, Piggin proposes a series of questions to analyse the political discourse applicable to sport policy, both in terms of policy production and dissemination [121].We can suggest that the aesthetics of the document produced may also play a significant role, e.g., in helping give meaning to actions.
Figure 1 summarises our conceptualisation attempt.

Discussion Conclusion: From an Analytical to an Action Framework for Researchers and Practitioners
Hoekman and Scheerder claimed that: "to pursue structural social change, more will be needed than merely publishing special issue and launching a debate concerning sport policy topics from an academic perspective" [11] (p. 104).We believe that this contribution may help to structure the debates on the definition of an NSF sport policy.
Our conceptualisation attempt has the merit of highlighting and linking the multiplicity of issues associated with the notion of NSF sport policy, concerning the formalisation process, the content of the policy and the transversality between structures within NSFs.Nevertheless, we are aware that any attempt at conceptualisation is not easy.Bloyce and Smith [122] (p.13) draw on Houlihan and White [2] to note the difficulty in conceptualising the term 'policy': All policies can be understood as involving overlapping and interrelated features: human action aimed at achieving certain goals [...] solving, or at least improving, an identified problem; and [...] maintaining or changing relationships within an organisation, between different organisations, or a human configuration of some other kind.
We have observed the predominance of public policy approaches.For sport organisations and a fortiori NSFs, we have seen that the issues are strategic (allocating resources according to guidelines, moving forward in concert with national and international contexts) and both intra-and inter-organisational.

Discussion Conclusion: From an Analytical to an Action Framework for Researchers and Practitioners
Hoekman and Scheerder claimed that: "to pursue structural social change, more will be needed than merely publishing special issue and launching a debate concerning sport policy topics from an academic perspective" [11] (p. 104).We believe that this contribution may help to structure the debates on the definition of an NSF sport policy.
Our conceptualisation attempt has the merit of highlighting and linking the multiplicity of issues associated with the notion of NSF sport policy, concerning the formalisation process, the content of the policy and the transversality between structures within NSFs.Nevertheless, we are aware that any attempt at conceptualisation is not easy.Bloyce and Smith [122] (p.13) draw on Houlihan and White [2] to note the difficulty in conceptualising the term 'policy': All policies can be understood as involving overlapping and interrelated features: human action aimed at achieving certain goals [...] solving, or at least improving, an identified problem; and [...] maintaining or changing relationships within an organisation, between different organisations, or a human configuration of some other kind.
We have observed the predominance of public policy approaches.For sport organisations and a fortiori NSFs, we have seen that the issues are strategic (allocating resources according to guidelines, moving forward in concert with national and international contexts) and both intra-and inter-organisational.
The proposed conceptualisation can be used both as an analysis model (see [123]) for researchers and as an action model for practitioners.
As a model of analysis, a double approach must be taken.On the one hand, there is a transversal analysis, by understanding the transversality between the macro, meso and micro levels.The aim would be to analyse the level of dissemination and appropriation of the sport policy between the different levels.On the other hand, there is a vertical analysis interested in specifically observing each level, verifying that the sport policy is well-formalised, -adapted and -implemented, and produces the desired effects for each of the macro, meso and micro levels.A challenge associated with this vertical analysis is the (sometimes necessary) transversality, with public policies at different levels.Whatever the approach, it is also possible to use this model to assess the success of an NSF sport policy, for example, by looking at each quadrant in Figure 1 and asking whether the policy addresses these issues, e.g., "are there any strategic orientations identified?"; "is a national-regional cooperation considered?"; and "does the NSF sport policy make sense at the local level?" As a model for action, there are a number of suggestions that can be made to practitioners.First, it is a question of giving indications for building a sport policy worthy of the name, i.e., a concerted strategic approach that is expressed in terms of objectives and action plans that each level of an NSF is able to implement.According to this model, the conditions for appropriation by the intermediate and local levels must be met.The issues are the involvement of stakeholders (governance, see [124]), communication and the alignment of resources with the ambitions expressed.There is, therefore, certainly a (political) process, but also a method to be put in place to involve stakeholders [101].In the formulation phase, Piggin et al., looking at the New Zealand sport policy, encourage "scholars and practitioners to conceive of policy as ongoing contests over truth.We [Piggin et al. (2009)] suggest that practitioners might benefit from considering the problematic implications of promoting "transparent" public policy" [125] (p.462).In our view, our conceptualisation should promote this search for truth and transparency by practitioners.
We acknowledge that our work has some limitations.From a methodological point of view, we have chosen to focus on the English and French literature, targeting titles and abstracts around keywords.Despite the snowballing strategy and the careful reading of significant books and articles, we are not immune from having missed other references, e.g., the most recent articles published after we stopped our search (e.g., [126][127][128]).Furthermore, each reference was only counted once (one stream, one field of study, one approach and one issue), while it is apparent that some articles are cross-cutting.Moreover, literature in languages other than English and French may have brought additional elements relevant to our conceptualisation.Future research could address these limitations.Other research and practical perspectives lie in the application of our conceptualisation to analyse an NSF sport policy or to guide an NSF in the development of its sport policy, with specific attention to the elements of the definition, issues and characteristics raised in the present paper.

3. 3 .
Towards a Conceptual Model to Better Understand an NSF Sport Policy Based on the above literature review, we can derive a conceptual model of what an NSF sport policy is and what issues this model should address for these organisations.

Figure 1 .
Figure 1.Conceptualisation attempt of an NSF sport policy.Source: authors.

Figure 1 .
Figure 1.Conceptualisation attempt of an NSF sport policy.Source: authors.

Table 1 .
Data analysis process.

Table 3 .
Summary of main topics, references and issues regarding studies about non-governmental sport policies.