The numbers game: quantifying good governance in sport

ABSTRACT Research question: The paper interrogates the growing quantification of good governance and its implications for sport organisations. It addresses what shapes the production of governance codes and who decides what aspects of governance to be quantified and how we govern and study the quantification of good governance in sport. Research methods This is a position and conceptual paper, which reviews three international and two national codes to illustrate the five dimensions of the quantification of governance, including the work required, its reactivity, its tendency to discipline human behaviour, its polyvalent authority, and its aesthetics. Results and findings The quantification of governance codes requires a significant amount of work and triggers compliance responses from sport organisations, which leads to structural changes and an ever-greater institutional complexity and need for capacity building. Quantification has challenged sport organisations’ autonomy as it does not promote self-governance but rather meta-governance. How the quantification of governance is related to its theorisation and codes’ accuracy validity remains problematic. Implications The paper challenges the uncritical acceptance of good governance codes as a norm for sport organisations and calls for a reconsideration of our approach in line with a sport-focused model, which acknowledges the value systems, local traditions and institutional frameworks underpinning the model.


Introduction
Codes of good governance (GGC) in sport have become a constitutive part of the profile of national and international sport organisations.The diffusion of GGC in sport has been promoted by coercive, normative and mimetic pressures exercised by intergovernmental, governmental, and non-governmental organisations (Geeraert, 2021;Girginov, 2021;Jack, 2018).The two main drivers of the diffusion of GGC in sport are efficiency and legitimation (McLeod & Shilbury, 2020;Pielke et al., 2019;Tacon & Walters, 2021), and mirror those in the corporate sector (Aguilera & Cuerro-Cazurra, 2004).Efficiency and legitimation are based on numbers expressed in governance numerical scores and percentages.Numerical expressions are central to quantification, a generic technology of government (Foucault, 1979).
Codes are a system of symbols used to represent assigned meanings, but they are not merely a neutral and apolitical description of reality.Codes are always representations of reality, and as their name suggests (from Latin 'codex' = law-book), they are designed to regulate people's and organisations' behaviours.GGC represents abstract social constructs, designed usually by experts and which utilise aggregate numbers to depict how an organisation is governed.Erkkilä et al. (2016, p. 319) explain that 'numbers expose problems, help institutionalise new domains of decision making, and make complex issues commensurable, giving them a common form.Numbers play a role in the construction of social reality'.Therefore, the quantification of GGC becomes a problem for sport organisations because it increases their institutional and policy complexity and serves not only to measure but to measure up or compare the performance of different organisations (Thévenot, 2019).Thus, there is a need to critically examine the challenges that the quantification of GGC poses to sport.
While the codification of good governance in sport has received some criticism (Geeraert, 2021;Parent et al., 2021;Walters & Tacon, 2018), virtually no attention has been afforded to three fundamental questions.Taking the lead from Berman and Hirschman (2018), this paper examines (i) what shapes the production of GGC in sport, and who decides what aspects of governance to be quantified; (ii) how the quantification of good governance in sport is governed; and (iii) how sport scholars study the quantification of governance, that is, how the construct of good governance is theorised concerning its quantification.Thus, this paper expands on the three knowledge gaps in the sport governance literature identified by Geeraert (2021), including its 'conceptual vagueness, unclear implementation rationale and the unclear impact of good governance strategies' (p.3).The renewed theoretical and practical interest in governance, as Jessop (2011) argues, is related to the simultaneous state and market failure to address the decline in social cohesion in the advanced capitalist societies and the resulting legitimacy and steering crisis and ungovernability.
This conceptual paper follows Gilson and Goldberg (2015) by taking a problem-focused approach and aims to develop a logical argument for associations between different dimensions of quantification rather than testing them empirically.The paper is structured in three parts.First, it interrogates the conceptualisation of good governance and its value for sport organisation.Second, it introduces a conceptual framework for understanding the quantification of good governance.Finally, the conceptual framework of quantification is applied to interrogate two national and three international GGCs in sport.

Origins and meaning of good governance
The first known corporate GGC emerged in 1978 (USA), followed by Hong Kong (1989), Ireland (1991) and the UK (1992) (Aguilera & Cuervo-Cazurra, 2009).Critically, the notion of good governance has always been evoked in the context of changing statesociety relationships.Thus, regardless of whether we think of governance as day-today management (e.g.organisational management); management of management (e.g.boards); management of corporate governance (e.g.dedicated authorities); governance of governance (e.g.laws, regulations), or meta-governance (e.g.constitutions, treaties) (Pérez, 2003), it always takes place in the context of state-society relations.For example, the 1337 painting 'Allegories of Good Governance' by the Italian master Lorenzetti was commissioned by the Siena to mark the demise of the absolute power of the monarchy and the emergence of a new type of city-states and a model of governance (Graaf & Asperen, 2016).The main factor leading to the adoption of the first American corporate code in 1978 was a series of corporate failures and the lack of any public oversight over the governance of big corporations (Price, 2018).The impetus behind the evolution of the UK good governance code in sport was the modernisation agenda of Blair's government of 1997 based on the tenets of New Public Management (NPM), which altered the state's role in sport (Houlihan & Green, 2009).A similar process has occurred in other countries (Andrews, 2018;McCree, 2009;Sam, 2009).
Governance is a political theory concerned with the relationship between state and society (Jessop, 2000;Stoker, 1998;Treib et al., 2007).According to Rhodes (1996, pp. 652-653), governance encapsulates 'a change in the meaning of government, referring to a new process of governing; or changed condition of ordered rule; or a new method by which society is governed'.
Good governance in sport is largely premised on the governance and management ideology of NPM, which is rooted in the neoliberal model of society and represents a reversal from the Keynesian social welfarism, and from managerial to entrepreneurial forms of governance (Harvey, 1989).It is explicitly concerned with efficiency, budget cuts, accountability for performance, audits, customer focus, strategic planning and management and competition (Gruening, 2001).The application of NPM in sport has been very problematic (Green & Houlihan, 2006;Lusted & O'Gorman, 2010).Yet, despite being long rejected as a model for public administration (Dunleavy et al., 2006;Hood & Jackson, 1992), the NPM ideology continues to shape GGC in sport.
The governance concept also faces significant linguistic and epistemological challenges.The word does not translate into many languages, and its key properties, such as the rule of law, transparency and effectiveness, have very different meanings across cultures.Karsten and Illa's (2005) analysis of traditional Western management models and Ubuntu (African) worldview illustrates the importance of cultural values and language.Similarly, Burger and Goslin (2005) argue that any contextual manifestation of good governance in sport should reflect the societal value system on which it is based and incorporate (South) African value systems.
Van Bottenburg (2021) points out two shortcomings of extant studies, including decontextualisation and de-historisation of sport governance, or the tendency to overlook the specific context and the relationship between state and sport organisations.Two European-wide studies reinforce his point.Chaker's (2004) survey of good governance in Europe reveals principal differences in state regulation of sport ranging from providing 25%-95% of National Governing Bodies' (NGB) funding.Breuer et al. (2015) show that most sport clubs in Europe pursue social objectives, and 24% of them are running a negative financial balance.On average, only 16% receive direct state subsidies.De Dycker's (2019) legal analysis of GGC in sport offers four useful criteria for understanding the relationship between the state and sport organisations.She distinguishes between (i) levels of state intervention; (ii) structure and responsibilities of the sport system; (iii) political responsibility for sport policy, and (iv) the organisational structure of the national sport movement.It follows that the GGC will perform different functions and will have a varied impact on different configurations of state-sport organisations relationships.Zattoni and Cuomo's (2008) analyse 60 national corporate governance codes based on common or civil law found that the issuance of GGC in civil law countries (i.e.France, Germany, Switzerland) is driven more by legitimation (institutional) reasons than by the determination to improve the governance practices of national companies.This finding challenges the conventional wisdom that good governance equates to effectiveness and brings the legal (i.e.politics) and institutional aspects of governance.Parent et al. (2018) also question the link between GGC and effectiveness in sport.The wider point, as Pieterse (2021, p. 6) explicates, is that 'Institutions derive from history, their backdrop is structural change whereas politics is situational.Governance arises from a confluence of institutions and politics'.
The epistemological, linguistic, cultural and institutional difficulties surrounding the construct of good governance suggest that its measurement will be equally challenging.As Thomas (2010, p. 7) points out, 'unlike measures that are simple count of events or objects, a measure of a theoretical construct is necessarily embedded in theory, which is what imbues the measure with meaning'.After all, quantifying is agreeing on what to measure and how and then measuring (Desrosières, 2011).Sport governance lacks adequate conceptualisation and has generally been taken to mean a set of principles to be followed by sport organisations.The uncritical insistence that promoting and measuring good governance enhances sport organisations' autonomy, democracy and effectiveness deserve scrutiny.The next section argues for why the ongoing quantification of GGC in sport poses serious challenges to sport organisations.

Conceptualising the quantification of good governance in sport
Numbers are critical for governing sport as they help make sense of time, distance, revenue, terms in office and board size.Quantification is a social process that becomes possible after conventions (i.e.how to categorise, respectively, how to measure) has been invented (Desrosières, 2011).Quantification entails expertise and knowledge and confers power, which allows the legitimate representatives of sport organisations to govern them by shaping issues and members' conduct.This section builds on the sociology of quantification framework developed by Espeland and Stevens (2008), who conceptualise quantification as 'social action that, akin to speech, can have multiple purposes and meanings' (p.495).Mennicken and Espeland (2019, p. 225) define quantification and commensuration as 'the comparison of different entities according to a common metrics'.They also argue that these purposes and meanings can only be understood in specific contexts.Espeland and Stevens offer five key dimensions of quantification, including (i) the work that it requires, (ii) its reactivity; (iii) its tendency to discipline human behaviour, (iv) its polyvalent authority, and (v) its aesthetics (p.493).
First, the work dimension entails that in all forms, quantification requires considerable intellectual investment from experts and policy-makers.As Espeland and Stevens (2008) put it 'rigorous, defensible and enduring systems of quantification require expertise, discipline, coordination and many kinds of resources, including time, money and political muscle ' (p. 411).Such systems, including those in sport, are usually produced by large bureaucratic organisations, such as the European Union (EU), the International Olympic Committee (IOC) and the Association of Summer Olympic International Federations (ASOIF).Therefore when imposed globally, they often create compliance problems and resentment (Geeraert, 2019).Second, reactivity refers to the capacity of measurement to intervene in the social reality it depicts.'Measures are reactive, they cause people to think and act differently' (Espeland & Stevens, 2008, p. 412).But measures possess another important characteristic concerning their ability to alter power relations because they affect how status, resources, knowledge and opportunities are distributed.Disciplining behaviour is both a key characteristic and output of quantification.It promotes certain practices that define what is appropriate, normal and aspirational and who should be subjected to what kinds of knowledge in an organisation.In the words of Espeland and Stevens, 'quantitative measures are a key mechanism for simplifying, classifying, comparing and evaluating that is at the heart of disciplinary power' (p.414).Third, quantification performs other disciplinary functions, including allowing various parties to check on people and organisations and make previously invisible behaviours and units visible.Fourth, for Porter (1995), the authority of numbers derives from their ability to resolve the fundamental problems of human communication across vast social, geographical and political distances.Espeland and Stevens extend this ability of numbers and suggest that their authority may be vested in our sense of their accuracy and validity and how they accumulate and connect users with a shared interest in numbers.Ultimately, the authority of numbers 'depends on establishing networks among objects and humans that become so sturdy they are no longer disputed or subject to disassembly' (p.421).National and international sport systems provide strong networks of organisations and members for asserting the authority of GGC.Finally, two widely accepted aesthetic criteria define numbers, including clarity and parsimony.GGC in sport has progressed in presenting the governance scores with greater clarity and legibility mainly by utilising infographics (e.g.ASOIF's report).GGC also tends to include only those indicators necessary and sufficient for communicating the information that matters to the users.While GGC scores may help us understand the abstract construct of governance through the beautiful graphics, we must not forget that such comprehension is always mediated by experts or those in power.Espeland and Stevens (2008) distinguish between two forms of quantification, including those that mark and those that are commensurate.The former provides useful knowledge, such as the number of club members and medals won, whereas the latter provides the valuation or meaning of different objects with a common metric.Espeland and Stevens raise another important point relevant to GGC in sport 'commensuration creates a specific type of relationship among objects.It transforms all differences into quantity.In doing so it unites objects by encompassing them under a shared common cognitive system' (p.498).Commensuration is articulated through statistics, including indicators, counts, ratios, ranking and intervals, which allows distinguishing between objects by 'assigning to each one a precise amount of something that is measurably different, or equal to, all others' (p.498).
Sport organisations' governance scores represent commensuration because they use common metrics to measure and compare levels of governance of different organisations.But before comparing sport organisations, they must be classified in ways that make them comparable.Berman and Hirschman (2018, p. 264) note that numbers perform a critical discursive function and require different interpretative work: generally, counts and ratios stay closer to the intent of their creators than indicators do because the former requires less interpretive work than composites and are less vulnerable to the twists produced by long interpretive chains.
Furthermore, the analysis of GGC borrows from Erkkilä et al.'s (2016) framework for classifying governance measurements.It accounts for the politics of governance measurement and suggests that all governance measurements involve fundamental underlying theoretical decisions and interpretations of social reality.The framework includes the type of producer organisation, purpose, scope, data gathering method, presentation of results and publication strategy and validity.As Demortain (2019) explicates, 'quantification has as its final step the 'extraction of a result'' (p.908), but there is a paucity of knowledge about the impact of quantification of GGC in sport.

Interrogating GGC in sport
To address the study's research questions, this section employs the quantification and measuring frameworks proposed by Espenald and Stevens and Erkkilla et al., respectively.These are used to examine three international GGC produced by ASOIF (2020), Sport 4 Good Governance adopted by the European Olympic Committees (European Olympic Committee, 2012), Sport Governance Observer developed by Play the Game (SGO, Geeraert, 2018) and two national codes developed by Sport Australia (ASC, 2020) and Sport England/UK Sport (2016).
ASOIF and EOC hold significant institutional power over their members.The two quasi non-governmental organisations, ASC and UK Sport, exercise a high level of control over 63 and 58 NGB, respectively.ASOIF is the umbrella organisation of 33 International Sport Federations (IF) and the EOC is the continental organisation of 50 NOC.The four organisations 1 have well-established networks of structures, members, officials, volunteers, sponsors and partners.
The GGC was analysed following Prior's (2008) two-pronged approach to the study of documents, including (i) those focusing on the content and (ii) on their use and function as well as whether the document is used as a resource or as a topic.This approach allows asking questions about what documents do with what they say.

Work involved in developing GGC
The development of all five GGCs required a significant amount of time, resources and political will, which are beyond the capacity of most sport organisations.For example, the ASOIF code was produced by an international task force group consisting of eight senior sport officials and one academic, while the S4GG and SGO have been designed mainly by academics in partnership with sport organisations, and as part of larger research projects funded by the EC's ERSMUS+ programme.Nonetheless, in all cases, a major political decision was needed on the part of the leading organisation, which involved complex consultations and negotiations with multiple stakeholders.
Tables 1 and 2 show the background characteristics of the five GGCs and the classification of measurements they use and allow drawing two conclusions.First, two forms of mimetic pressures were at play.All GGCs have been modelled on the existing corporate governance codes and the political ideology of the NPM.The national codes have been refined over a sustained periodfrom the beginning of the 2000s in the UK and from 2012 in Australia.The ASOIF, S4GG and SGO codes were developed in response to mimetic pressures chiefly from the IOC and EU, respectively.
Second, all GGC codes use commensurate measures, which produce degrees of good governancefrom categories A1, A2, B, and C (ASOIF) to four 25 percentage categories of compliance (EOC), average scores (ASC, SGO) and three levels of compliance (UK).The use of the different number of loosely defined indicators by GGC (from 50 -ASOIF to 274 -SGO) requires a great deal of interpretation.Rankings, on the other hand, are a zero-sum game where one organisation's raise in the ranking requires another organisation's descent.As Girginov (2008) notes, sport development is a labelling game, which is explicitly defined, and cannot be conceived of without a notion of its opposite of underdevelopment.Like 'development', 'good governance' serves as a constructor because it can create its opposite of bad governance, and carries specific power.Thus, GGC scores not only tell us who the best-governed sport organisations are but also 'name and shame' those that have failed short of the expected standards.The commensurable power of GGC is questionable, though, because except for the ASOIF's and UK's codes, which use only the size of the organisation, as measured by its budget and number of staff, the rest of the codes do not make these organisations comparable.Moreover, the GGCs (save SGO) have been backed by experienced sport bureaucracies capable of monitoring and correcting their implementation.The remainder of this section follows the other four dimensions of quantification (Espeland & Stevens, 2008) to answer the main research questions.

GGC as a reactivity measure
There are virtually no studies on organisational reactions triggered by the implementation of GGC apart from research examining compliance.Tacon andWalters' (2016, 2021) investigation of the implementation of the UK GGC reveals several challenges, including the lack of organisational capacity and a top-down imposition of standards against the culture and traditions of NGB.Parent et al. ( 2021) report similar implementation issues with Canadian NGB, and Geeraert (2019) notes that IF's compliance is contingent on capacity.He identifies four sources of non-compliance, including monitoring and sanctioning, persuasion and social learning, transfer of knowledge and resources and rule interpretation, which undermine the universal adoption of GGC.Sport organisations' reactions to the implementation of GGC indicate their capacity to shape the social reality of governance on the ground.The governance action plans of several UK NGB demonstrate that some structural changes in the composition of the boards and in instituting new commissions and policies had been made.The reactivity of sport organisations is largely determined by one of the key principles of GGCtransparency.Intended or not, all commensurate measures rely on transparency to promote reactivity.By publishing the results of the implementation of GGC, the producers of the codes hope to highlight good and unsuccessful practices.But transparency is a double-edged sword, and as Berman and Hirschman (2018) note, 'Many numbers are meant to incentivize certain behaviour, and if their calculation is mysterious, they can't produce the desired effects.As long as numbers have consequences, though, people will try to game them; and transparency facilitates this' (p.264).Since most GGCs rely on self-evaluations, there will always be the possibility to game them.

GGC as a disciplining instrument
The disciplining power of GGC deserves particular attention because it concerns the very foundation of sport organisations or their autonomy.Autonomy is a multifaceted concept which may take several forms (Mrkonjic, 2016).The quantification of governance serves to reinforce discipline in the sense that it promotes certain organisational behaviours and an outcome in that it produces new social realities.One of these new realities concerns the ability of third parties to check on sport organisations' performance, as demonstrated by the growing power of athletes within the Olympic Movement (Chappelet, 2021a).
Sport autonomy, according to Olympic Charter (IOC, 2020), is the right to self-governance free from external controlling influence.This is a recognition that sport organisations are complex open systems that historically have been premised on their ability to self-govern themselves (Chappelet, 2021b).From a systems theory's perspective, most sport organisations are autopoietic systems, the core idea of which is the notion of self-governance (Chelladurai, 2014).As Esmark (2011) observes, 'the concept of autopoiesis basically suggests that a system is its own output: an autopoietic system produces itself rather than something else, the latter being the defining trademark of allopoietic systems' (p.93).Unlike the typical business, which produces goods and services consumed by geographically dispersed customers, sport organisations are the producers and consumers of their services (Chelladurai, 2014;Hoye, 2007).Although the range of sport services on offer has gradually expanded, the relationship between their production and consumption has not changed.
Autopoietic systems, such as sport, always maintain a boundary with their environment, which is a critical precondition for their autonomy.As a result, they display a significant mistrust of any attempt to steer from the outside.What follows is that autopoietic systems do not instantly react to external stimuli rather, they respond according to their logics, traditions and values as determined by the system itself.At the same time, autopoietic systems always maintain a cognitive openness to external influences and use it to enhance their self-governing capabilities.Esmark (2011) explicates that 'The key proposition of the autopoietic paradigm is rather that steering from the outside always involves steering of self-steering, or governance of self-governance' (p.94).Thus, governance and self-governance exist in a relationship where self-governance of a system constitutes first-order governance and attempts to govern the system from the outside are considered second-order governance (i.e.meta-governance).The critical question then becomes to what extent strategies to impose governance of sport organisations from the outside promote and strengthen their self-governance.As Jessop (2003) explicates, the concept of meta-governance is based on Ashby's 'law of requisite variety', which suggests that the steering of a particular system always requires the capacity to match the complexity of that system.As the history of sport suggests, no government possesses this capacity (Hoberman, 1984), but what some governments have done is to change the meaning of autonomy.According to the Governance Institute (2019): The introduction of the new Code for Sports Governance, and its requirement for mandatory compliance, is a bold move, representing definitively the end of autonomy, at least for funded organisations in the UK, and the introduction in its place of 'earned autonomy.(p. 5) None of the GGC reviewed mentions the notion of autonomy 2 , which reinforces the critical question of to what extent these codes promote self-governance or the autonomy of sport organisations.One can, however, infer the relationship between self-governance and governance from the intent of GGC to build the capacity of NGB.These have been designed to support strengthening sport organisations' structures and processes (i.e.capacity).Some codes include explicit measures (e.g.ASC, S4GG, UK).In the case of the UK NGB, there is a dedicated fund of £2 million to support staff training, technology and processes development.The ultimate purpose of this support package is to ensure greater ownership and self-regulations of organisations within the newly established 'earned' autonomy.
The ASOIF's (2020) report explicitly recognizes organisational capacity as the key determinant of governance score, but the Association offers very little support to its members.GGC, thus, actively nurture the three core processes responsible for promoting disciplined conduct, including training individuals in various routines, putting them under panoptic surveillance, and punishing them for proscribed or deviate behaviour (Foucault, 1979).GGC envision training sport officials in certain routines, a regular gathering of information on various aspects and classifying monitored organisations according to predetermined categories.In this way, the GGC turns into a tool that can reveal and control problematic conduct.

Authority of GGCcommunication, accuracy and validity
The authority of any GGC is contingent on its accuracy and validity, so we can judge how governance numbers communicate and connect sport policy-makers and stakeholders across national and international spaces.The raison d'etre of GGC is to support the development of sport, but as far as it can be ascertained, no GGC has been questioned on the normative and policy prescriptions they have exerted on the sport development agenda.Measuring concerns a relationship between the subject and object of investigation, and its purpose is inference.It has been defined as 'any process by which a value is assigned to the level or state of some quality of an object of study' (Bulmer, 2001, p. 455).Despite claims contrary, measurement is never a neutral, objective and technical concept aiming at capturing an already existing and independent reality.Instead, as Fineman (2004) points out, measurement does not reflect but constitutes social reality, thus the need to scrutinise its accuracy and validity.Malito's (2015) extensive analysis of the challenges in measuring governance raises the critical question of 'how policy demands, and normative prescriptions inform the construction of indicators'.She concludes 'The study of indicators of governance has been limited to find descriptive solutions that have rarely attempted to explore the normative reasons guiding the numerical representation of political phenomena' (p.21).
Andrews, Russell and Barrios' (2016) examine the relationship between governance ends and means in development through sport in 80 national and regional governments echoes Malito's concern.They identify three common governance ends, including inclusion, growth and health but caution that the mix of what might be called "ends" and "means" in a single number leads to a loss of information about goals and tools.It could lead to some less-than-optimal results where means drive ends and not the other way around.(p.21) The five governance codes' indicators reviewed represent paymaster serving concepts as they are overly normative and subscribe to an abstract model of promoting good governance.Indicators ought to reflect the mission, values and strategic objectives of the organisation.As Greer et al. ( 2016) note, we judge the nature and quality of any governance measures by how they fulfil three fundamental roles, including an accountability measure, a measuring instrument and a research instrument.These roles are critical for assuring sport participants and stakeholders that sport organisations have sound structures and policies in place that guarantee the quality and safety of their services (i.e.accountability); organisational performance is in line with the stated mission and strategic objectives (i.e.management); and measurements are used to identify governance arrangements promoting best practices and capable of rectifying issues (i.e.research).
The GGCs reviewed partly fulfil the first fundamental role of measuring governance in that they capture some structural and processual aspects, such as the composition of boards and the presence of election policies.Nonetheless, they largely neglect other important issues concerning the quality, range and accessibility of service provision.While all GGCs allocate a score for the presence of a mission statement, none of them captures the alignment of mission statements with overall organisational performance and developmental goals.Table 3

illustrates the point.
There is a notable lack of alignment between the GGC issuer organisation's mission and most measuring indicators unless one engages in an elaborate interpretive exercise, which automatically undermines the accuracy and validity of the indicators.Regarding the research instrument's role in governance measures, the scores they provide offer some insights into the presence of certain structures and policies.Nonetheless, these indicators are largely ineffective for research purposes because they fail to address the much more important how and why questions of governance.Numbers have agency, write Mennecken and Salais (2021), 'because numbers order and make possible specific kinds of cognition and action and preclude others' (vii).
The ultimate test for any measuring indicator is its relation to the abstract construct it represents (i.e. level of abstraction and conceptualisation), its construct validity; the hypothesised relationship with the main governance principle/standard, the multi or uni-dimensionality of data used and the functional relationship between indicators.Governance is an inherently unobservable construct, and to be adequately measured, it requires first to establish its relation to a particular theorisation, which will provide the definition of the concept to be operationalised.Second, the proposed governance definition must be unpacked and turned into several observable proxy variables, which are then used to derive a measure of the construct.Except for the SGO, the rest of the GGC do not seem to follow this established scientific logic, which undermines their meaningfulness as a true representation of the good governance concept.Furthermore, no GGC reports on its construct validity or any type of errors (i.e.content, convergent or criterion validity).
The lack of accuracy and validity of governance measures raises four main issues.First, all proposed measures represent composite indicators.These indicators are particularly attractive to sport officials because they reduce complex governance realities to a single number (a 'point score'), but they hide important issues such as the role of sport-specific inequality, access to sport and power relations.Second, all good governance indicators rely on the elite (i.e.expert) and not on public evaluation.As Girginov and Preuss (2021) argue public value represents a relationship between a valued object (i.e.governance) and a valuing subject (i.e. the public); that is, public value is in the eye of the beholder, so it is critical to define who the public is.Sport organisations have several publics, including staff, athletes, members and coaches, among others.No GGC has been based on public evaluation (this is not to be confused with stakeholders' consultation) and Norris (2011) examination of governance codes demonstrates a significant divergence between elite and public evaluations.Yet, the ultimate purpose of good governance is to create public value (Jørgensen & Sørensen, 2012).Third, there is a remarkable similarity between the measures of all governance codes.All codes use multi-dimensional and aggregate measures, which blurs the distinction between the interpretation of the governance concept at national and international levels.Finally, the GGC examined to create a transparency paradox (Oman & Arndt, 2010).While they purport to promote transparency, how the proposed governance measures have not been fully transparent.
Following Thomas (2010) well-substantiated critique of governance indicators, some measure-specific questions can be raised.For example, what is the hypothesised relationship between the key governance principle of the presence of mission, vision and strategy (S4GG/EOC) and the indicator 'risk management', or between the principles of 'democracy' and 'integrity' and the indicators 'clear policy ensuring candidates opportunity to campaign' and 'appropriate gender balance in Executive Board', respectively (ASOIF)?What do 'democracy and 'integrity' mean, and would the risk management indicator be better put in the cluster 'accountability, transparency and control'?It is also worth asking about the functional relationship between indicators because all governance principles are mutually constructive and interrelated.How is then possible for an IF to score high on democracy and transparency and low on control mechanism and development simultaneously (ASOIF, 2020)?

Aesthetics of GGCclarity and parsimony
Espeland and Stevens (2008) argue that 'despite variations, two aesthetic ideals for numerical pictures are virtually universal: clarity and parsimony' (p.421).Clarity of numbers is expressed in easily legible information, while parsimony is manifested by communicating only the necessary and sufficient information.As discussed above, governance measures result from a logical chain of elements, the first of which is the governance theory.The Sage Handbook of Governance (Bevir, 2011) discusses 10 competing governance theories, from institutional to network, system and meta governance.None of the GGC reviewed references any theory nor provides a conceptualisation of good governance, with the partial exception of Play the Game (Geeraert, 2018) and ASC.Geeraert acknowledges the polity (i.e.state-society relations) and politics (i.e.power constellation) dimensions of governance but the SGO chooses to interpret governance as a policy instrument, which focuses on understanding what actions have been taken to ensure the democracy, transparency and solidarity of sport organisations.A similar reductionist approach is part of the problem, as one cannot isolate governance from the context in which it takes place, that is, polity and politics.Sport organisations are complex open systems, and we ought to understand their ability to self-govern themselves and resist environmental pressures, including top-down imposed models of governance.
The GGC examined does not provide a clear definition of governance (i.e.lack of clarity).While they use fewer principles (i.e.4-9, parsimony), the operationalisation of those principles varies from 37 (ASC) to 274 (EOC) indicators, thus raising many questions about the communication of critical information about governance.A related point concerns the lack of clarity between input and output indicators.Input indicators refer to the governance structures and processes and are supposed to answer the question of how good governance is implemented in sport organisations.Output indicators concern the consequences of good governance in all aspects of the organisation.With the partial exception of the ASOIF's code, which makes a passing reference to the impact of term limits on the overall score of IF, the rest of the GGC does not include any output indicators.Nonetheless, to claim any causal relationship between imposing office terms and higher overall governance score would be far-fetched.The ASC's GGC acknowledges 'Good governance does not guarantee success; however, poor governance almost certainly guarantees failure' (ASC, n.d.).

Conclusions
This paper addresses the questions what shapes the production of GGC in sport, and who decides what aspects of governance to be quantified, how the quantification of good governance in sport is governed, or how numbers shape the organisation of sport, and how sport scholars study the quantification of governance.It located these questions two broader issues concerning the challenges of using aggregate numbers for understanding the complexities of modern sport organisations and how the use of numbers alters previously established structures and practices.
Sport governance studies suggest that the proliferation of GGC has been driven by models borrowed from the corporate sector.Nonetheless, market forces have been actively complemented by government interventions and turned into policies.The key aspects of good governance subject to quantification have been accountability, transparency and democracy, but integrity and solidarity were also operationalised.Sport's uncritical borrowing from the tenets of NPM has been very problematic.In responding to Chalip's question of whether to build sport management on a sport-focused model or a model deriving from mainstream management, Gammelsaeter (2020) writes 'my answer is that academic theories and concepts are not neutral.If we build sport management science on theories derived from industry, we should not be surprised if sport is conceptualised as industry or kin to industry' (p.8).
The five GGCs reviewed show a great deal of similarity although they have been developed for very different constituencies and contexts.Thus, their quantification has helped shape the bahaviour of sport organisations in line with a vaguely defined notion of good governance, which has a much greater political appeal than practical relevance.Governance codes' convergence across national contexts has become widespread (McLeod et al., 2021;Van Eekeren, 2021), which has raised more questions than the implementation of the codes was supposed to tackle.
The quantification of GGC in sport requires a significant amount of work and resources, which most organisations do not have.As a reactive measure, GGC triggers compliance responses from sport organisations, which leads to structural changes and an ever-greater need for capacity building.Taken as an instrument of discipline, GGC has challenged the autonomy of sport organisation as they do not necessarily promote self-governance but rather meta-governance or the imposition of organisational standards from outside.The accuracy and validity of GGC have been highly problematic as these have failed to account for a range of important aspects of sport organisations' work.The clarity and parsimony dimension of GGC remains uncertain.Despite recent advances in sport governance literature, scholars have failed to address the issue of how the quantification of governance is related to its theorisation.Equally important remains the question of the impact of GGC on sport organisations' power relations, autonomy and day-to-day management, to which, so far, we have no clear understanding.
The introduction of GGC in sport is primarily a governance issue.Thus, it is ironic that we promote governance codes but have no clear idea which governs them and how they are altering the institution of sport.This paper does not propagate abandoning good governance but rather the need for its critical reconsideration in line with a sportfocused model, which acknowledges the value systems, local traditions and institutional frameworks underpinning the model.

Notes
1. Play the game only loosely fits in this category because of its status as a research institute.2. Save two passing references in the ASC code in relation to the autonomy of CEO to make decisions.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Table 1 .
Background characteristics of selected good governance codes (GGC) in sport.

Table 2 .
Classification of good governance codes' measurements.

Table 3 .
Relationship between sport organisations' mission, sport development goals and governance indicators.