Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Safeguarding, Power, and Organizational Culture in Sport

Abstract. Safeguarding in sport is an organizational system of prevention, recognition, disclosure, protection, accountability, and repair, shaped by unequal power and culturally sanctioned performance demands. This integrative research monograph examines how coaching authority, selection dependency, institutional incentives, peer norms, reporting climates, and governance structures influence interpersonal violence and athletes’ ability to seek remedy. Historical sport research, prevalence studies, qualitative organizational investigations, systematic reviews, and the 2024 International Olympic Committee consensus are evaluated alongside general organizational theories. The evidence consistently identifies multiple forms of maltreatment and barriers to disclosure, while robust causal evaluation of prevention programs remains limited. The synthesis proposes operationally testable organizational indicators, multilevel evaluation strategies, and a research agenda for trustworthy safeguarding across sport settings.

Keywords: sport safeguarding; interpersonal violence in sport; athlete welfare; organizational culture; coach–athlete power; psychological safety; abuse disclosure; institutional betrayal; sports governance; safe sport; organizational silence; preventive evaluation

Safeguarding as an Organizational Research Object

Safeguarding in sport denotes the organized capacities through which participants are protected from interpersonal violence, exploitative conduct, neglect, and retaliation while retaining safe access to participation, support, and remedy. The object of this dossier is the social and institutional system that produces these outcomes: authority relations, selection rules, routine interactions, shared expectations, reporting channels, independent oversight, and responses to allegations. An effective analysis identifies the conditions under which maltreatment can occur, remain concealed, be challenged, and be repaired. The 2024 International Olympic Committee consensus situates interpersonal violence in a socioecological environment and treats responsibility as distributed across the sport ecosystem Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). That orientation makes organizational processes empirically relevant in their own right, beyond attributes of any individual perpetrator or athlete.

Organizational safeguarding is observable at several points in the life of a concern. Prevention concerns whether risky practices are recognized and corrected before harm occurs. Disclosure concerns whether athletes can communicate difficulties without sacrificing their sporting future. Response concerns whether reports generate timely protective action, fair investigation, appropriate referral, and documented accountability. Recovery concerns whether affected individuals regain access to support and belonging. Each stage depends on trust, power, institutional incentives, and practical resources. A policy document may describe all four stages while daily operations fail at one or more of them. The difference between policy existence and actual institutional behavior is central to research Johansson, S (2022).

The unit of analysis must remain explicit. An athlete's experience of humiliation is an individual report of an interpersonal event. A team in which humiliation is routine presents a shared climate. A federation that rewards abusive coaches despite formal prohibitions presents a governance problem. These phenomena interact, yet each requires distinct evidence. This monograph examines those relations using sport-specific surveys, qualitative interviews, systematic syntheses, international consensus work, and carefully identified organizational theory. It treats proposed safeguards as interventions requiring evaluation, with particular attention to the limited quantity of strong causal prevention evidence Roberts, V., Sojo, V., & Grant, F (2020) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Conceptual Boundaries and Operational Definitions

Interpersonal violence in sport covers acts and omissions that create psychological, physical, or sexual harm within relationships and institutions associated with organized sporting activity. The language of maltreatment commonly includes psychological abuse, physical abuse, sexual abuse, and neglect; peer aggression, harassment, bullying, hazing, and discriminatory treatment may overlap with those categories. The IOC's 2016 statement broadened a previously prominent sexual-harassment focus to include psychological abuse, physical abuse, and neglect, and its 2024 statement developed a still broader ecological understanding Mountjoy, M., Brackenridge, C., Arrington, M., Blauwet, C (2016) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). A sound study defines each exposure behaviorally, specifies its time frame, and reports whether a single incident, repeated pattern, perceived threat, or independently assessed case qualifies.

Power means asymmetric control over resources, opportunities, legitimacy, or decisions that matter to the less powerful party. In sport, such resources include roster selection, scholarship renewal, competition access, training time, medical referrals, team belonging, and career recommendations. Organizational culture refers to shared assumptions and enacted patterns of meaning; organizational climate refers more immediately to participants' perceptions of expected and rewarded conduct. Neither construct can be adequately inferred from a glossy mission statement. They are measured through repeated practices, convergent independent accounts, staff and athlete surveys, decision records, and the treatment of people who challenge dominant norms Roberts, V., Sojo, V., & Grant, F (2020) Morrison, E. W., & Milliken, F. J (2000).

Psychological safety is a team-level belief that speaking up with questions or concerns is interpersonally permissible. It has established roots in organizational learning research Edmondson, A (1999). Safeguarding requires additional provisions, including confidential disclosure options, protection against retaliation, independent investigation, child protection procedures, and practical support. A team might encourage technical dissent while suppressing allegations against a celebrated coach. Conversely, a formal reporting mechanism might exist in a setting where athletes believe that use of the mechanism will end their careers. Psychological safety is therefore a relevant mechanism and measurement domain; it is not an adequate substitute for safeguarding outcome indicators.

Evidence Sources, Review Logic, and Inferential Boundaries

This monograph is a critical integrative narrative synthesis. Its evidence base was constructed through targeted inspection of original sport studies, sport-specific systematic reviews, official consensus and position statements, and selected foundational organizational research. It does not assert that all eligible studies were located and does not report a PRISMA search, independently duplicated full screening, risk-of-bias scoring across every study, or a quantitative meta-analysis. These distinctions matter because the literature combines retrospective surveys, athlete narratives, organizational case studies, professional recommendations, and relatively few preventive intervention evaluations. Different designs answer different questions; treating their conclusions as interchangeable would conceal the central limitations of the field Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Original empirical reports anchor claims about what investigators observed. The survey by Vertommen and colleagues asked more than four thousand adults in the Netherlands and Belgium about childhood experiences in organized sport Vertommen, T., Schipper-van Veldhoven, N., Wouters, K., Kamp (2016). The Canadian national-team study surveyed 995 current or recently retired athletes about maltreatment Willson, E., Kerr, G., Stirling, A., & Buono, S (2022). Johansson's federation study combined 18 interviews with review of the public material of ten Swedish sports federations Johansson, S (2022). Parent analyzed organizational documents and interviewed 27 sport stakeholders about sexual-abuse disclosure Parent, S (2011). These studies provide complementary evidence about exposure, institutional practice, and response, but their sampling frames and observational methods set limits on causal generalization.

The research also uses a systematic review of organizational factors Roberts, V., Sojo, V., & Grant, F (2020), the 2024 IOC expert consensus derived from an extensive scoping review and Delphi process Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024), and a systematic review published on October 28, 2026 addressing sexual violence, its correlates, dynamics, and possible disruption of sporting belonging Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). Formal consensus identifies shared expert judgments and practical priorities; it cannot turn an unevaluated intervention into a demonstrated remedy. Cross-domain organizational theory helps specify mechanisms such as silence and institutional betrayal but requires sport-specific validation before its parameters are treated as established in athlete populations Morrison, E. W., & Milliken, F. J (2000) Smith, C. P., & Freyd, J. J (2014).

Historical Development of Safe Sport Scholarship

Research on harassment and sexual exploitation in sport developed through feminist sport sociology, studies of athlete authority relations, and investigations of institutions in which athletic success could outweigh individual welfare. Brackenridge's historical work helped make the sexual vulnerability of female athletes a recognized problem in sports medicine and sporting governance Brackenridge (2000)70198-8). Fasting and colleagues subsequently compared 660 Norwegian elite female athletes with age-matched nonathletes and reported greater harassment by male authority figures among athletes, despite a lack of overall group difference in sexual harassment or abuse prevalence Fasting, K., Brackenridge, C., & Sundgot-Borgen, J (2003). The distinction was scientifically important: context and perpetrator authority could matter even where aggregate rates were similar.

The field expanded from identifying sexual violations to mapping emotional and physical harm. Stirling and Kerr analyzed interviews with 14 retired elite female swimmers to categorize emotional abuse through physical behavior, verbal behavior, and withdrawal of attention or support Stirling, A. E., & Kerr, G. A (2008). That qualitative evidence clarified how coaching practices can convey rejection and domination without the legal or behavioral markers used to classify sexual abuse. The IOC consensus of 2016 brought psychological, physical, and neglect-related harms into a wider athlete-protection framework Mountjoy, M., Brackenridge, C., Arrington, M., Blauwet, C (2016). These developments shifted the research question toward the everyday organization of training, dependence, and acceptable conduct.

A later institutional turn investigated implementation, disclosure, and governance. Parent's work examined why disclosure is socially and organizationally difficult Parent, S (2011); Rhind and colleagues analyzed experiences from a safeguarding initiative involving 32 sport organizations Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F (2017); Johansson documented gaps between policies and daily federation practice Johansson, S (2022). The 2024 IOC consensus formalized a whole-system conceptual model and 43 agreed recommendations, and the 2025 FEPSAC position statement emphasized the contribution of sport psychology to safeguarding Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024) Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025). The 2026 synthesis of sexual violence proposed the hypothesis of a dual injury to the person and to sporting belonging Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). This trajectory reveals increasing methodological and conceptual attention to institutions as active determinants of risk and response.

The Socioecological Organization of Risk

A socioecological model organizes determinants across interacting levels. At the individual level, developmental stage, social position, access to knowledge, dependence, and previous experiences may shape exposure and reporting. At the interpersonal level, coaching relationships, peer norms, grooming, intimidation, and bystander conduct affect immediate safety. At the organizational level, personnel screening, supervision, incentive systems, selection rules, and investigation structures influence opportunities for harm and redress. At the broader sociocultural level, gender hierarchies, public prestige, legal norms, funding regimes, and expectations about obedience and toughness influence organizational choices Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). A complete causal hypothesis specifies the mechanism through which one level alters another.

The model is especially valuable for avoiding a false assumption that individual resilience can compensate for an institution's unsafe design. Teaching athletes to identify abusive behavior may improve recognition, but their ability to act remains constrained if reporting threatens selection or if alleged perpetrators control the only available investigation channel. Similarly, training coaches about professional conduct may fail to change the behavior of a federation that consistently rewards humiliating coaching methods. Roberts and colleagues identified power imbalance, winner-take-all reward arrangements, physical or psychological isolation, organizational tolerance, and role conflict among important categories in the literature Roberts, V., Sojo, V., & Grant, F (2020). The relationship of these categories to actual incidents requires temporally informative, multi-source empirical testing.

A whole-system model also exposes the risk of ecological inference. A sport with a strong performance culture may contain both harmful and protective teams. An athlete may report abuse without knowing which governance mechanisms caused it. Organization-level correlations cannot automatically establish individual causation, and individual anecdotes do not quantify prevalence. Hierarchical sampling across clubs, teams, age groups, and federations can identify how context predicts risk while appropriately accounting for clustering. Repeated measurement and documented policy changes can show whether the proposed mechanisms behave as expected. The strength of the socioecological approach lies in its capacity to generate these identifiable cross-level tests rather than in the model's breadth alone.

Power, Dependency, and Control of Sporting Opportunity

Coaching authority combines legitimate educational expertise with control over scarce opportunities. Selection decisions, competition entries, recommendations, travel permissions, and access to influential networks can become instruments of conditional acceptance. Athlete dependence grows when pathways are narrow, replacement is easy, and competing sources of instruction are unavailable. In that setting, conduct that would be questioned elsewhere may be interpreted as the price of advancement. The organizational review by Roberts and colleagues identifies power imbalance and isolation as important structural conditions linked in the literature to non-accidental violence Roberts, V., Sojo, V., & Grant, F (2020). These mechanisms are propositions about organizational risk, not a presumption that expertise or competitive hierarchy is intrinsically abusive.

Resource dependence is intensified by uncertainty. An athlete cannot always know why a coach withdrew approval, excluded them from a squad, or changed their training load; discretion may be professionally legitimate. The absence of transparent criteria makes retaliation harder to identify and creates incentives for self-censorship. This problem is especially severe when a person who can affect selection also receives a complaint, evaluates the complainant's credibility, and determines the complainant's subsequent opportunities. A useful safeguard separates contested decision-making roles, documents selection standards, and creates genuinely independent review. Research should measure athletes' anticipated career consequences of speaking up, comparing teams with different reporting and selection arrangements Parent, S (2011) Willson, E., Kerr, G., Battaglia, A., & Stirling, A (2022).

Power also operates through affection and belonging. A coach's withdrawal of support can function as punishment without a formal sanction, while a team's social network can enforce silence after a concern is raised. Stirling and Kerr's interviews identified denial of attention and support as one form of emotionally abusive behavior Stirling, A. E., & Kerr, G. A (2008). The causal role of such practices depends on context, repetition, intention where it can be established, the athlete's circumstances, and observable effects. Researchers must distinguish ordinary performance feedback from humiliation, coercive threats, or strategic deprivation of support through behaviorally detailed assessment. The relevant organizational question is whether athletes retain meaningful alternatives and protections when authority is misused.

Winner-Take-All Incentives and Performance Legitimacy

Sport organizations commonly pursue rankings, medals, selection, sponsorship, and public recognition under high uncertainty. These goals can create a reward structure in which a successful coach's methods receive less scrutiny and apparent results become a substitute for ethical evaluation. Roberts and colleagues identified winner-take-all reward systems as a structural category with potential relevance to non-accidental violence Roberts, V., Sojo, V., & Grant, F (2020). A plausible mechanism is selective tolerance: leaders excuse boundary violations when the person responsible seems difficult to replace. A related mechanism is normalization: athletes learn that distress or humiliation is accepted as evidence of commitment. Both mechanisms require evidence beyond the mere presence of competitive ambition.

A performance-centered culture is created through routine decisions. If staff promotion depends almost exclusively on short-term wins, safeguarding information has limited influence over reward allocation. If athletes who disclose concerns are characterized as uncommitted, the organization creates a deterrent to future disclosures. If winning coaches are investigated by their own colleagues, conflicts of interest can transform nominal procedures into protective shields for status. These possibilities connect formal performance incentives to everyday relational norms. The account of organizational silence developed by Morrison and Milliken offers a general framework for how powerful leaders and prevailing beliefs inhibit upward communication Morrison, E. W., & Milliken, F. J (2000). Its specific causal pathways should be examined with sport data.

Outcome evaluation should distinguish competitive performance, coaching effectiveness, retention, athlete welfare, and protection from violence. Strong performance in one domain cannot validate a harmful practice in another. The empirical question whether coercive methods improve results requires controlled comparison, temporal attention to short- and long-term outcomes, and measurement of harm. Retrospective statements that an athlete succeeded under an abusive coach do not establish that the abuse caused the success. Survivorship and selection can make the most visible successful athletes unrepresentative of those who withdrew, were injured, or never entered elite pathways. Longitudinal tracking across the entire athlete pipeline would provide a more defensible view of welfare and performance jointly.

Organizational Culture, Climate, and the Normalization of Harm

Culture becomes visible in what staff and athletes can do repeatedly without challenge, in stories told about good performance, and in the response to violations by high-status members. Formal policy signals an intended standard; enacted culture is expressed in selection, supervision, social sanctions, and the distribution of attention. An organization's climate includes athletes' expectations about whether concerns will be heard, whether confidentiality will be respected, and whether safeguarding is genuinely rewarded. Johansson's investigation of Swedish federations showed how explicit rejection of sexual abuse could coexist with limited operational measures and a tendency to keep the issue invisible Johansson, S (2022). The study directly illustrates why a safeguarding audit must examine implementation rather than only policy language.

Normalization proceeds through social learning and coordination. Junior athletes observe how experienced participants speak about intense coaching, injury, privacy, and boundaries; repeated transgressions may become ordinary. A practice can remain invisible to formal leadership because it occurs during travel, informal training, or private communication. Isolation and poor communication are organizational risk categories identified in the literature Roberts, V., Sojo, V., & Grant, F (2020). A cultural explanation should specify measurable features such as routine inappropriate remarks, observed intervention by peers, patterns of unchecked complaints, and perceived consequences for reporting. Simply labeling a team 'toxic' provides little guidance for intervention or causal identification.

Culture can be investigated comparatively through confidential climate surveys, observation, document analysis, and independently collected interviews across organizational ranks. Researchers should test whether athletes, parents, volunteers, coaches, and safeguarding personnel share the same perceptions; disagreement can itself be diagnostic of unequal information or vulnerability. Time series following leadership changes can reveal whether policies are translated into behavior. The central analytical question is how an organization converts values into routine, enforceable practices and whether those practices remain stable when competitive pressure rises. A mature safeguarding culture is supported by feedback mechanisms able to detect and correct failures rather than by reputational claims alone.

Coaching Authority, Boundaries, and Emotional Abuse

Coaching requires instruction, evaluation, correction, and at times direct intervention in physical technique. Ethical authority depends on clearly bounded purposes, proportional means, transparency, and the athlete's continued dignity. Emotional abuse is studied through recurrent behaviors such as degradation, intimidation, rejection, and the withdrawal of essential support. In their qualitative inquiry involving 14 retired elite female swimmers, Stirling and Kerr found categories of verbal, physical, and attention-related abusive conduct Stirling, A. E., & Kerr, G. A (2008). Such results identify meaningful behavior patterns but do not estimate population prevalence. Research must assess frequency, severity, perceived control, cumulative exposure, and whether participants could challenge the behavior without losing their sporting position.

Interpretation often becomes complicated when participants believe harsh treatment is causally necessary for success. Athletes may retrospectively connect demanding coaching with achievement; coaches may describe coercive practices as discipline. Causal evaluation has to distinguish the impact of technically appropriate difficult training from the impact of humiliation, threatening language, and conditional interpersonal acceptance. Interviews with athletes can establish the meaning and context of conduct, but retrospective narratives alone cannot determine whether a harmful intervention improved performance. The literature on athletes' perceptions of abusive coaching relationships makes the dependency of the athlete-coach bond especially salient Stirling, A. E., & Kerr, G. A (2009).

Organizational prevention requires observable boundaries: instruction and selection criteria that can be explained, supervised training arrangements, conduct standards applying to high-status personnel, independent complaint pathways, and procedures for separating performance decisions from grievance handling. Training must include scenarios where ostensibly motivational language masks threats or degrading control. Evaluation should examine whether ordinary coaching corrections remain robust while harmful conduct declines. This turns the vague demand for respectful culture into measurable practices and counterfactual questions. Strong sport science protects both the legitimacy of coaching expertise and the athlete's ability to refuse abusive treatment without a predictable career penalty Roberts, V., Sojo, V., & Grant, F (2020) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Sexual Violence, Grooming, and Boundary Transgression

Sexual violence in sport includes unwanted sexual contact, coercion, harassment, and other forms of sexual exploitation linked to sporting environments. Grooming may involve progressive trust building, privileged attention, secrecy, boundary testing, and manipulation of personal or professional dependence. It can occur through direct authority relationships or through peers and other personnel; a systematic review published in 2026 stresses the diversity of perpetrators and the relevance of multiple-offender configurations Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). Research should distinguish sexual violence from consensual relationships, while recognizing that ostensibly consensual interactions under strong asymmetrical dependence require careful contextual and legal examination.

The historical comparison of Norwegian elite female athletes and nonathletes found more harassment by male authority figures within the athlete sample despite no corresponding overall difference in harassment or abuse prevalence Fasting, K., Brackenridge, C., & Sundgot-Borgen, J (2003). That finding illustrates why risk is not adequately summarized by one aggregate number. Authority changes access, credibility, and the perceived cost of disclosure. Victims may be reliant on the individual they would otherwise report and may fear exclusion from a pathway developed over many years. Institutional knowledge of this dependency creates a responsibility to provide independent recourse, meaningful safeguarding education, and restrictions on unmonitored high-risk situations Parent, S (2011).

Prevention cannot rely solely on the ability of athletes to identify or resist grooming. Structured supervision of training and travel, clear professional boundaries, privacy-aware communication policies, accessible reporting, and separation of investigation from the alleged offender's managerial network are organizational components worth testing. Their effectiveness must be established through meaningful outcomes rather than policy compliance alone. Moreover, safeguarding responses should preserve procedural fairness, confidentiality, and the welfare of complainants and witnesses. The IOC's whole-system recommendations support multi-layer protective design, while leaving intervention-specific causal effects an open empirical question Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Physical Violence, Neglect, and the Limits of Toughness

Physical abuse includes intentional harmful force or coercive physical practices outside legitimate sporting technique. Neglect concerns failures to provide necessary care or protection where duty and capacity are present; examples considered in sport may include ignoring known injury risks, withholding essential support, or disregarding an athlete's developmental needs. The 2016 IOC consensus identified physical abuse and neglect as crucial categories alongside sexual and psychological violence Mountjoy, M., Brackenridge, C., Arrington, M., Blauwet, C (2016). Definitions require care because strenuous exercise, physical contact, and high training loads are normal in many sports. The investigator should examine purpose, consent or lawful authorization, proportionality, conditions of performance, medical guidance, and possible coercion.

A culture of extreme toughness can reduce the visibility of harmful practices when pain, fear, or exhaustion is automatically treated as proof of dedication. Athletes may have limited ability to distinguish acceptable physical demands from unsafe orders if they depend on the same personnel for opportunity and technical evaluation. Such ambiguity becomes organizational when medical opinions are overridden, injuries are concealed, or staff are rewarded for short-term performance despite repeated warning signs. These are hypotheses about governance and incentive structures, requiring triangulation of medical records where ethically accessible, training documentation, athlete reports, and contemporaneous decisions Roberts, V., Sojo, V., & Grant, F (2020).

Safeguarding research should not collapse injury occurrence into abuse prevalence. A sports injury may occur despite appropriate prevention, while an abusive order can exist without a recorded injury. Separate measurements are needed for violent conduct, omission of duty, injury, medical decision-making, athlete pressure, and organizational response. The question of organizational preventability requires attention to foreseeability and available protective options, rather than merely bad outcomes. Multi-level review can identify recurring patterns of harmful practice even when no individual episode contains all information necessary for classification Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Peer Aggression, Hazing, Bullying, and Bystander Dynamics

Abusive conduct in sport is not confined to coaches. Peers may engage in bullying, discrimination, initiation rituals, coercive group practices, threats, and sexually harmful behavior. A team can create a norm in which informal leaders enforce compliance and new members absorb humiliation as a condition of belonging. The 2024 IOC consensus addresses interpersonal violence across the ecosystem and recognizes perpetrators beyond a single authority role Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). The 2026 review of sexual violence identifies peers and other sporting personnel among relevant perpetrators Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). A comprehensive assessment must therefore examine horizontal as well as hierarchical relations.

Peer-inflicted harm may remain undetected because the behavior occurs away from staff, is described as banter, or is normalized by participants seeking acceptance. Individual disclosure decisions are shaped by expected solidarity and retaliation. Bystanders may have information but fear ostracism or believe intervention will be ineffective. Team-level climate research can estimate whether members perceive the conduct as legitimate, whether intervention is socially supported, and whether leaders consistently act when concerns arise. Such indicators should be compared with independently recorded incidents and follow-up outcomes to reduce the risk that socially desirable answers obscure persistent problems.

An effective response architecture establishes consistent boundaries for initiations, shared living spaces, travel, digital team groups, and athlete-led activities. Education should specify observable conduct and support witnesses who raise concerns. Research designs can test whether peer norms change after carefully implemented interventions and whether reported harms change after accounting for shifts in willingness to disclose. Increases in reporting following improved outreach may indicate greater trust and detection rather than rising victimization. Valid evaluation therefore requires repeated behavioral, climate, and response-quality measures rather than a single incident count Willson, E., Kerr, G., Battaglia, A., & Stirling, A (2022).

Disability, Gender, Sexuality, and Unequal Protection

Exposure to interpersonal violence and the ability to report it may vary with disability, gender, sexuality, social status, and other intersecting circumstances. The IOC's 2024 review describes differences in the available evidence across categories and cautions that findings are uneven; for some variables, studies produce inconsistent estimates Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). The Dutch and Belgian retrospective survey reported higher recalled exposure among several groups, including respondents identifying as disabled, ethnic-minority, or lesbian, gay, or bisexual Vertommen, T., Schipper-van Veldhoven, N., Wouters, K., Kamp (2016). Such results warrant targeted attention without justifying a universal risk ranking across all sports or societies.

Some athletes rely on assistance with personal care, transportation, communication, specialized equipment, or classification procedures. Additional dependencies may expand opportunities for abuse when boundary definitions are unclear or external oversight is absent. Female and gender-diverse athletes may face sexualized conduct or reputational sanctions, while racialized or migrant participants may experience language barriers and institutional distrust. These are plausible mechanisms that require population-specific research. Safeguarding standards should therefore be accessible across language, communication mode, disability accommodation, and cultural context, with the athlete able to choose a trusted route to support.

Methodological equity involves sampling participants whom conventional anonymous online surveys may miss. Researchers should evaluate whether instruments are comprehensible, whether behavioral descriptions translate accurately, and whether disclosure through accessible systems creates privacy risks. Small subgroup cells, self-selection, and the risk of identification can weaken published estimates. A careful study reports denominators, missing data, uncertainty, and the definition of the subgroup before drawing conclusions. Institutional inclusion is demonstrated through practical access to protection, independent assistance, and fair treatment of concerns, as well as through representative participation in designing safeguards Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025).

Elite Pipelines, Scholarship Dependency, and Career Retention

Elite sport can magnify the consequences of challenging authority because competitive places are scarce and years of investment depend on selection or institutional patronage. A complaint may affect contracts, scholarships, access to international competitions, reputation, or the perceived future of a whole support team. In a Canadian study, 995 current and recently retired national-team athletes reported maltreatment experiences, providing a substantive but context-specific picture of elite-sport exposure Willson, E., Kerr, G., Stirling, A., & Buono, S (2022). Separate qualitative work with athletes asked what changes would advance safe sport, highlighting the value of participants' own accounts of institutional barriers and possible solutions Willson, E., Kerr, G., Battaglia, A., & Stirling, A (2022).

Career dependence complicates apparent consent and silence. An athlete may remain in an abusive environment because alternative training arrangements are unavailable; another may leave without filing a report. Studies restricted to current teams can therefore omit people whose harmful experiences led to withdrawal. Retired athlete samples supply additional perspectives but introduce recall bias and changes in norms over time. Assessments should deliberately recruit current athletes, recent leavers, those excluded from programs, and support staff, while distinguishing voluntary exit from removal. The research question is whether protection is consistent across selection outcomes and stages of career.

Sport organizations can reduce coercive dependence by publishing selection criteria, documenting appeals, separating welfare services from selection decisions, and preserving independent pathways to medical, legal, or psychological consultation. These interventions are institutionally plausible but require formal evaluation. A well-designed study would compare reporting trust and adverse career outcomes before and after a governance reform, while assessing whether reforms were applied consistently and whether athlete composition changed. The key outcome is the athlete's practical freedom to contest misconduct without an institutionally imposed cost that discourages truthful disclosure Roberts, V., Sojo, V., & Grant, F (2020) Parent, S (2011).

Prevalence Research and the Problem of Denominators

Prevalence estimates are inseparable from definitions, sampling, and recall windows. The Dutch and Belgian survey found that 38% of adult respondents with childhood organized-sport participation reported psychological violence, 11% physical violence, and 14% sexual violence during childhood sport experiences Vertommen, T., Schipper-van Veldhoven, N., Wouters, K., Kamp (2016). These are estimates for the surveyed retrospective population and the study's classification rules. They should not be reported as an annual incidence rate, a universal world prevalence, or the proportion of current children in every sport. Differences in age, level of competition, national context, and recruitment can change the apparent distribution of experiences.

The 2024 IOC synthesis documents exceptionally wide study-level ranges for interpersonal and sexual violence, reflecting broad heterogeneity in operationalization Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). The 2026 systematic review of sexual violence included 63 publications representing 58 independent studies and reported study-level prevalence below 2% up to 61%, with a higher estimate in a particular female subgroup Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). Such dispersion is a warning against averaging incompatible studies into a single simple number. A study measuring one act ever experienced is answering a different question from a study measuring repeated severe acts within the previous year or allegations established through formal investigation.

Prevalence studies must report participant eligibility, sampling mechanism, exposure to organized sport, definitions of violence, period at risk, denominator selection, nonresponse, and uncertainty. Athletes who have already left sport following harm may be underrepresented in current-club surveys. Self-report can reveal experiences absent from formal records while also being shaped by memory, interpretation, survey mode, and trust. Official complaints capture people who entered a reporting system, not all people harmed. For substantive evaluation, prevalence estimates should be compared within matched definitions and settings, with separate measurement of reporting behavior and response mechanisms Willson, E., Kerr, G., Stirling, A., & Buono, S (2022).

Consequences, Belonging, and Recovery After Harm

Interpersonal violence can have psychological, social, physical, and career-related consequences, including withdrawal from teams, loss of trust, disruption of identity, and reduced willingness to participate. The pathways are heterogeneous: an athlete's reaction depends on the nature and duration of exposure, available support, institutional response, and other life conditions. The 2026 systematic review of sexual violence synthesizes adverse outcomes and proposes a sport-specific dual-harm hypothesis in which victimization may coincide with disruption of belonging and participation Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). That formulation offers a testable interpretation of the literature; it is a hypothesis rather than a clinically established syndrome.

Organizational reaction can shape consequences. An institution may respond promptly and provide practical support, or it may publicly discredit a complainant, delay investigation, deny access to records, or force continued exposure to the person named in the report. Institutional-betrayal research outside sport shows how perceived institutional failures can be associated with additional trauma-related difficulties Smith, C. P., & Freyd, J. J (2013) Smith, C. P., & Freyd, J. J (2014). These findings justify examining betrayal as a possible explanatory construct in athletic settings, while direct estimates should come from sport-specific longitudinal research. The harm of a breached relationship with a sporting institution can be experienced as a loss of community and trusted structure.

Recovery must be studied beyond the final outcome of an investigation. Important indicators include perceived safety, continuity of care, control over participation choices, restoration of social connections, procedural fairness, and freedom from retaliation. Researchers should avoid assuming that returning to the same team is always desirable or that leaving sport proves a poor response. Athletes may define a successful outcome as safe exit, a change of coach, a fair hearing, or renewed access to sport without forced contact. Trauma- and violence-informed responses emphasize safety, choice, and dignity while recognizing that clinical assessment and treatment require qualified professionals when indicated Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Disclosure, Silence, and Anticipated Retaliation

Disclosure is a relational and organizational process in which an athlete evaluates whether speaking will produce assistance, dismissal, retaliation, or loss of belonging. Parent's study of sport organizations combined policy analysis with interviews involving 27 stakeholders and examined how organizational contexts influence sexual-abuse disclosure Parent, S (2011). This evidence brings attention to trust, access, norms, and response rather than framing disclosure solely as a matter of the individual's courage. A participant may recognize harmful conduct yet reasonably believe that there is no safe person to tell or that disclosure will jeopardize essential sporting opportunities.

Organizational silence can be produced by patterns of authority that discourage upward information, by defensive routines that protect reputation, and by shared assumptions that leaders prefer not to hear criticism Morrison, E. W., & Milliken, F. J (2000). In sport, those mechanisms can coincide with selection power and an expectation of loyalty to the team. Retaliation may be direct, such as exclusion or threats, or indirect, such as withdrawn support and reputational labeling. A rigorous safeguarding survey should measure anticipated and experienced retaliation separately and ask whether specific actions preceded or followed disclosures, while protecting participants from inadvertent identification.

Reporting counts are ambiguous. A new confidential route may initially increase reported concerns because people trust it more; a campaign to suppress disclosures may decrease counts despite worsening conditions. For evaluation, organizations need separate measures of exposure, awareness of channels, willingness to disclose, volume of reports, timeliness, perceived fairness, retaliation, and independent outcomes. Examining each stage allows researchers to distinguish actual prevention from altered detection. Athlete-centered qualitative research supplies indispensable insight into why formal channels are or are not considered usable Willson, E., Kerr, G., Battaglia, A., & Stirling, A (2022).

Institutional Betrayal, Credibility, and Secondary Injury

An institution may cause additional injury when it fails to prevent foreseeable harm or responds in ways that violate participants' reasonable reliance on its protection. Smith and Freyd developed the concept of institutional betrayal through research into trusted organizations and survivors' experiences Smith, C. P., & Freyd, J. J (2014). Their 2013 study of 345 college women found associations between reported institutional betrayal surrounding unwanted sexual experiences and several adverse psychological measures Smith, C. P., & Freyd, J. J (2013). The study took place outside sport, so its estimates are not sport-specific. Its relevance here is theoretical: the response of a trusted institution is a possible component of exposure and recovery rather than a purely administrative afterthought.

Credibility assessments are structurally vulnerable when an athlete's account challenges a respected coach, major sponsor, or governing body's public image. The risk is heightened when investigators share strong professional ties with the person implicated or when decision-makers hold implicit beliefs about which athletes are reliable. Procedural harm can arise from unnecessary repetition of interviews, unmanaged confidentiality, public speculation, or failure to explain the process. Studying these possibilities requires careful distinctions among allegations, investigative findings, institutional actions, and affected participants' own evaluations. Ethical research should preserve due process while examining how differences in power influence evidence collection and access to support.

Independent review, documented conflicts of interest, clear communication, and access to supportive services may reduce the opportunities for institutional betrayal, but their preventive effects need testing. A useful outcome battery would include time to protective response, informed consent in information sharing, perceived impartiality, continuity of services, absence of retaliation, and long-term participation decisions. Comparisons across governance arrangements must account for differences in incident severity and reporting rates. A system that simply concludes more cases quickly could still fail athletes if its procedures deny them meaningful voice or fair protection Parent, S (2011) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Independent Reporting and Investigative Architecture

A credible reporting route must be accessible to participants who have reasons to distrust ordinary management. It needs sufficient operational independence from coaches, selection panels, team executives, and parties whose reputations depend on the outcome. Independence has practical dimensions: funding, reporting lines, authority to preserve evidence, power to refer matters to statutory authorities where required, conflict-of-interest controls, and oversight of delay. The IOC's 2024 consensus addresses integrated safeguards and response capacities across sport organizations Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). The literature supports the importance of credible arrangements, but comparative estimates of which precise architecture produces the best outcomes are relatively scarce.

Confidentiality and anonymity require accurate promises. A confidential report protects information within specified rules; a truly anonymous report limits what can be known about the source but may restrict the ability to investigate. Some allegations trigger child-protection, criminal, or professional reporting obligations that vary by jurisdiction. Safeguarding staff must communicate these limits before soliciting disclosures where practicable. Data minimization, secure storage, controlled access, retention schedules, and transparent escalation procedures are essential in design. Research should test whether participants understand the process and whether institutional practice corresponds to its stated assurances.

Independence also matters after an initial report. Investigative decisions should be documented, affected parties should have appropriate opportunities to contribute evidence, and interim protective measures should be proportionate and reviewed. Athletes require access to information about their own welfare and procedural next steps without exposure to unnecessary confidential material. Researchers can compare organizations with different degrees of external oversight using quasi-experimental methods, although confounding by organizational maturity and resources will be substantial. An important distinction separates the quality of a safeguarding system from the proportion of cases resulting in disciplinary sanction; fair investigations may yield different outcomes without implying systemic failure.

Procedural Justice, Fairness, and Due Process

Procedural justice concerns how people evaluate the fairness, neutrality, consistency, transparency, and respectful treatment embodied in decision-making processes. In sport safeguarding, these qualities affect both a complainant's willingness to use a mechanism and the legitimacy of organizational action. An independent investigator can still undermine trust through unexplained delay, confusing standards, or dismissive communication. Conversely, a process can support participants even when available evidence does not establish a specific allegation. Fairness therefore requires attention to the dignity and rights of complainants, respondents, and witnesses throughout a difficult inquiry, alongside lawful and proportionate protective measures.

Research should distinguish perceived fairness from investigative accuracy. A participant may feel heard and still disagree with a finding; a technically sound evidentiary process may nevertheless generate distress. Valid measurement collects different dimensions, such as clarity of information, opportunity to provide evidence, impartiality, timeliness, confidentiality, retaliation, protective action, and eventual adjudicative outcome. Aggregation into one satisfaction score can conceal serious failures. Multi-informant audit and records-based process tracing are needed to compare what occurred with what procedures required. The broader organizational literature on silence suggests that trust in leadership affects whether concerns become visible in the first place Morrison, E. W., & Milliken, F. J (2000).

Due process also provides a scientific safeguard against interpretive overreach. Studies of allegations must label self-reported experiences, confirmed findings, disciplinary decisions, and legal judgments distinctly. A report is important evidence of an experience and a reason for protective attention, while causal and legal conclusions require their own methods and standards. Conflicting accounts and missing records should be handled transparently rather than collapsed into a binary measure of institutional success. Policy evaluation needs endpoints that capture both protection and the integrity of the investigation, avoiding incentives to maximize sanctions as an administrative performance target Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Psychological Safety, Athlete Voice, and Legitimate Dissent

Organizational research defines psychological safety in terms of a shared perception that interpersonal risk-taking, such as asking questions or admitting errors, is possible without predictable humiliation. Edmondson's study of work teams established an influential framework connecting psychological safety with team learning behavior Edmondson, A (1999). The proposed transfer to sport is analytically useful because athletes must communicate information about injuries, coaching problems, unsafe facilities, and perceived abuse. Yet speaking up about a dangerous training load involves different risks from accusing a powerful coach of sexual misconduct. Sport-specific measures should distinguish everyday voice from high-stakes safeguarding disclosure.

Athlete voice concerns opportunities to influence decisions, introduce concerns, challenge norms, and participate in the rules governing participation. It depends on actual organizational response, not merely the presence of an athlete committee. Consultation is limited when representatives can attend meetings but cannot obtain records, secure independent advice, or shape disciplinary and selection rules. Research with Canadian national-team athletes emphasizes the significance of athletes' own perspectives on reform Willson, E., Kerr, G., Battaglia, A., & Stirling, A (2022). Participatory design can improve the relevance of safeguarding mechanisms; rigorous evaluation still needs to establish whether participation changes behavior and outcomes.

Investigators can operationalize psychological safety through anonymous repeated team surveys, speaking-up behavior, feedback follow-through, and whether dissenting participants subsequently lose opportunities. Combining these observations with independent incident data helps evaluate the credibility of self-reports. Organizations must guard against a methodological trap in which people with the strongest reasons to fear retaliation are least likely to complete climate surveys. Allowing secure off-site responses, measuring nonresponse, and comparing athlete and staff accounts provide partial protection. The ultimate question is whether legitimate dissent can occur under real competitive pressure and when the complaint concerns the person controlling selection Roberts, V., Sojo, V., & Grant, F (2020).

Trustworthy Coaching Leadership and Accountability

Effective coaching includes technical expertise, motivational communication, individualized feedback, and decisions that may be unpopular. Safeguarding requires those functions to be accountable to professional boundaries and clear responsibilities. A coach should be able to explain the purpose and proportionality of demanding tasks, while an athlete should have routes for evaluating concerns outside the same authority relation. Emotional-abuse studies show that conduct can take the form of verbal degradation and withdrawal of support, which might otherwise be mistaken for ordinary performance management Stirling, A. E., & Kerr, G. A (2008). Organizational standards must examine the enacted behavior rather than the coach's self-description alone.

Leadership accountability depends on what institutions reward and correct. Screening and induction are insufficient when abusive behavior is excused after sporting success. Performance appraisals should contain credible indicators of athlete welfare, adherence to boundaries, and responsible response to concerns, while preserving confidentiality and fairness. Managers need training to identify conflicts of interest, respond appropriately to disclosures, and refer allegations rather than conducting improvised investigations. An evaluable program would track uptake of training, observed practice, supervisor responses, documented complaints, and follow-up outcomes. The distinction between policy adherence and prevention efficacy remains essential Johansson, S (2022).

A protective coaching environment should support athlete autonomy and meaningful participation within legitimate sport requirements. It can include supervision, regular independent welfare contact, transparent selection criteria, clear communication about one-to-one sessions, and reviewable training decisions. However, prescribing a universal supervision format across all sports, ages, and disability settings may overlook important practical differences. Research should identify core safeguarding functions and test context-appropriate implementations. The IOC and FEPSAC statements offer international guidance on broad principles while empirical program evaluation remains necessary to show which arrangements achieve measurable improvements Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024) Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025).

Roles of Families, Medical Teams, and Sporting Entourages

Sport safeguarding responsibilities extend beyond the athlete and head coach. Parents, guardians, medical teams, psychologists, welfare officers, club administrators, travel staff, volunteers, and national federations each possess different information and authority. Distributed responsibility can be protective when concerns are recognized early and passed to competent professionals. It can also become fragmented when everyone assumes that someone else has acted. A whole-system model therefore requires role clarity: who receives concerns, who provides immediate protection, who preserves confidential records, who arranges independent investigation, and who coordinates support Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Families play different roles depending on age, independence, and the athlete's circumstances. Guardians often provide essential protection for minors, but safeguarding cannot assume every family relationship is a safe disclosure channel. Medical and psychological professionals may notice injuries, distress, or patterns of intimidation that competition staff overlook; their professional obligations and confidentiality constraints must be respected. The 2025 FEPSAC position statement explicitly brings applied sport psychology into safeguarding discourse Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025). Clinical evaluation, when needed, belongs to qualified professionals operating under appropriate professional and legal frameworks.

Coordination should be measured as an observable organizational process. For example, an independent case review may assess whether a welfare concern reached the correct safeguarding professional, whether an interim protective measure was documented, whether privacy obligations were explained, and whether medical advice was separated from selection pressure. Research can examine bottlenecks and conflicting incentives across the entourage. High-quality collaboration does not require indiscriminate sharing of sensitive data; it requires defined consent and lawful reporting rules, minimum necessary information, clear responsibility, and records adequate to verify that promised protections occurred Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F (2017).

Safeguarding Governance, Resources, and Organizational Learning

Governance determines whether protective procedures survive the ordinary pressures of selection, budgets, staff turnover, and reputational crisis. Essential elements may include board-level responsibility, independent oversight, trained safeguarding personnel, explicit delegation, adequately funded support services, audit of high-risk settings, and meaningful athlete participation. In their examination of safeguarding development across 32 organizations, Rhind and colleagues described eight CHILDREN pillars: cultural sensitivity, holistic approaches, incentives, leadership, dynamic processes, resources, engagement, and networks Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F (2017). Those pillars provide a practical organizational framework derived from qualitative pilot evidence; they do not establish the effect size of any particular intervention.

Resource allocation is a direct test of institutional commitment. A federation may publish high standards while leaving staff too little time to investigate, failing to finance independent advice, or restricting access to support outside metropolitan centers. Governance must therefore be evaluated through decisions and outputs rather than statement counts. The federation study by Johansson is particularly instructive because representatives affirmed the unacceptability of sexual abuse even while dedicated measures remained limited Johansson, S (2022). Such gaps can be studied through implementation fidelity, staffing records, response timeliness, budget stability, and athlete awareness.

An organizational-learning system collects information about failures, investigates underlying mechanisms, changes procedures, and checks whether the change persists. Learning should include anonymized review of near misses, patterns of repeat complaints, high-risk transitions, and unintended consequences of formal controls. It also needs protection against overconfidence when a policy appears comprehensive. The IOC review notes the need to measure prevention and response effectiveness across the sport ecosystem Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). A mature governance system welcomes credible external audit and treats unfavorable findings as information required for correction rather than as threats to institutional identity.

Prevention Across Recruitment, Training, Travel, and Transitions

Safeguarding risk varies by environment and activity. Recruitment creates access to authority positions; residential camps and travel can reduce ordinary supervision; one-to-one training and medical contact introduce privacy and boundary questions; selection periods heighten dependence; transitions between clubs can interrupt continuity of records. A preventive strategy maps these contexts and establishes safeguards matched to actual risk. Screening, codes of conduct, informed boundary-setting, supervision, safe transport arrangements, and escalation routes are plausible design components, but their individual contributions to reducing victimization require evaluation Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Implementation should specify what actions staff must take, how those actions can be audited, and which protections are available when routine systems fail. The strongest plan can be undermined by exceptions for star personnel or by procedures that exist only at national headquarters. Consistency across local clubs, community programs, youth development centers, and elite national teams is a difficult coordination problem. The organizational review of non-accidental violence suggests attention to isolation, resource dependence, and communication stressors when assessing structural exposure Roberts, V., Sojo, V., & Grant, F (2020). One universal checklist cannot replace contextual analysis of how power is actually exercised.

Prevention also requires anticipating displacement. Restricting one unsupervised setting may shift harmful interactions to private messaging or off-site relationships unless the broader boundary and reporting system functions. Introducing a complaint channel without reliable protection against retaliation may increase exposure to institutional harm. Consequently, multicomponent strategies should be evaluated for behavioral change, unintended effects, accessibility, and sustained implementation over time. Researchers should compare policy-defined compliance with independently elicited athlete experience and objective process indicators, being careful not to mistake increased disclosure for evidence that the intervention created additional violence Willson, E., Kerr, G., Battaglia, A., & Stirling, A (2022).

What Safeguarding Interventions Have Actually Demonstrated

The most important evidence gap concerns causal effectiveness. Safeguarding policies, training sessions, background checks, independent reporting services, and case-management systems are widespread candidates for protection, but their existence does not establish that interpersonal violence has decreased. The 2024 IOC consensus emerged from an extensive review and Delphi process and emphasizes both implementation and measurement of effectiveness Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). Organizational research documents barriers and promising structures, while comparatively few studies isolate the effects of prevention programs on valid, behaviorally defined outcomes. A fundamental monograph must distinguish well-supported evidence that abuse occurs from weaker evidence about which particular combinations of safeguards prevent it.

Rhind and colleagues' work on the international child-safeguarding initiative involved qualitative evidence from 32 organizations and identified the CHILDREN organizational pillars Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F (2017). That design can inform feasibility, acceptability, implementation mechanisms, and contextual barriers. It cannot provide a randomized estimate of reductions in victimization. Johansson's investigation of Swedish federations described inconsistent policy translation and measures that were often at an early stage Johansson, S (2022). The study supports a diagnosis of implementation problems, but not the claim that any specific reform will reliably solve them. The FEPSAC statement supplies professional guidance grounded in contemporary literature, not a new controlled intervention trial Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025).

Future evaluations need to register the intervention components, the comparison condition, exposure duration, and a small set of valid primary outcomes before collecting data. Cluster randomization may be appropriate where organizations can be assigned to an enhanced safeguarding program without withdrawing basic legally and ethically required protection. Where randomization is infeasible, phased rollout, interrupted time series, or matched comparison designs can improve inference if secular trends, organizational differences, and reporting changes are addressed. Every study should report possible harms of an intervention, including unintended exposure of complainants or increased informal exclusion. The criterion is reliable reduction of risk and improvement in protective response, not the size of the training audience.

Safeguarding Education, Awareness, and Behavioral Transfer

Education can convey definitions, recognizable scenarios, routes to assistance, boundaries of professional contact, and obligations of staff and participants. These are legitimate intermediate goals because people cannot reliably use a policy they do not understand. The unresolved scientific question is whether changed knowledge persists, transfers into ordinary coaching and peer interactions, and ultimately reduces harmful conduct. Many safeguarding evaluations rely on satisfaction or post-training awareness rather than observed behavioral changes or independent measures of harm Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). A robust assessment must specify the pathway from training content to protective action and identify the conditions under which that pathway could fail.

Training designs should be sensitive to the audience. A minor athlete needs age-appropriate explanations of safe and unsafe conduct and multiple trusted reporting options. A coach needs practical boundaries and examples of harmful practices disguised as motivation. A board member needs understanding of conflicts of interest, protection from retaliation, and the duty to fund impartial procedures. A safeguarding officer needs competence in trauma-informed communication, documentation, information security, and appropriate referral. These differing requirements make a universal one-hour training metric an inadequate measure of organizational preparedness Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025).

Behavioral-transfer evaluation can combine scenario decisions, delayed knowledge checks, observed staff practices, anonymous athlete reports, and process audits following hypothetical or authentic concerns. Researchers should compare cohorts with appropriate baseline measures and avoid assuming that respondents who learned more are those most likely to challenge a powerful authority figure. Organizational incentives remain part of the mechanism: an individual trained to intervene may be prevented from doing so by managerial norms. The most credible program tests both skill acquisition and the availability of real institutional conditions for using those skills Roberts, V., Sojo, V., & Grant, F (2020) Morrison, E. W., & Milliken, F. J (2000).

Measuring Maltreatment and Organizational Exposure

Measurement begins by separating the exposure from its consequences. Behaviorally specific items can assess threats, humiliation, unwanted touching, coercion, discriminatory treatment, forced isolation, or denial of necessary care. Separate items can record frequency, recency, setting, perpetrator relationship, and perceived freedom to refuse. An athlete's current distress, withdrawal from sport, or fear of reporting is scientifically important but should not be used as an interchangeable indicator of whether a particular harmful act occurred. The heterogeneous prevalence findings assembled by the IOC and the 2026 systematic review show why construct breadth and recall period must be made explicit Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024) Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026).

Organizational exposure can be measured at multiple levels. Individual measures cover dependency, perceived retaliation risk, and trust in reporting. Team measures cover perceived norms, interpersonal voice, and the prevalence of tolerated violations. Organization-level indicators can include independence of investigations, safeguarding staff resources, response times, staff training records, board oversight, and whether substantiated conduct changes personnel decisions. Models must account for the fact that these measurements have different causal and statistical meanings. A document that says a policy exists describes formal adoption; it does not establish whether athletes know it, use it, or receive protection Johansson, S (2022).

Instrument validation requires cognitive interviewing, reliability analysis, longitudinal stability where appropriate, and evidence that measures distinguish closely related constructs. For example, a team may score high on ordinary psychological safety while scoring poorly on safe disclosure of abuse by senior staff. Researchers should examine discriminant validity rather than simply combining all favorable perceptions into a 'safe culture' factor. Measurement invariance across sport, age, gender, disability, and language settings is essential before comparing group means. Data collection must protect respondents against identification when a small team or a rare event makes anonymity fragile.

Retrospective Surveys, Missingness, and Disclosure Bias

Retrospective surveys offer access to experiences that may never have entered official incident systems. Vertommen and colleagues recruited adults with childhood organized-sport experience and obtained a large dataset covering psychological, physical, and sexual violence Vertommen, T., Schipper-van Veldhoven, N., Wouters, K., Kamp (2016). Such methods can reveal patterns across long sporting histories, including those who later left sport, but recall can be altered by time, changing language, and reinterpretation of past events. Online participation also raises the question of which former athletes were motivated and able to respond. These limitations should be reported alongside prevalence estimates rather than treated as a reason to disregard the participants' experiences.

Current-athlete surveys answer a different question. The Canadian national-team study involving 995 participants offered a substantial view of maltreatment among current and recently retired elite athletes Willson, E., Kerr, G., Stirling, A., & Buono, S (2022). Yet elite pathways selectively retain people who remain involved or reachable through institutional networks. Some with severe experiences may withdraw early, while others may decline the survey because confidentiality feels uncertain. Institutional surveys can also encounter social desirability and fear. Researchers should use independent recruitment channels where possible and disclose response rates, missing data patterns, eligibility criteria, and protections against coaches viewing individual answers.

Anonymous surveys and official case records should be triangulated rather than treated as competing estimates of the same quantity. An official case count reflects opportunities, incentives, and procedures for reporting as well as underlying events. A self-report survey reflects the wording, access, and memory of the sampled population. Longitudinal designs with repeated confidential measures may help detect changes, but participants can become recognizable as time series accumulate. Statistical models for missingness should be accompanied by sensitivity analyses because missing data may depend on precisely the harms being investigated. Reporting uncertainty is a core requirement of trustworthy safeguarding science.

Multilevel Models and Organizational Causal Identification

Sport participants are nested and cross-classified within teams, coaches, clubs, federations, countries, and career periods. A single athlete may change coaches or participate in multiple institutions, while one coach may influence many athletes. These structures require analysis that recognizes clustering, temporal order, and nonindependence. A multilevel study can estimate whether team climate and institution-level policies predict individual exposure after relevant variables are considered. However, observed association cannot establish causality where safer organizations differ in unmeasured recruitment, resources, or broader legal protections. The organizational factor synthesis offers substantive hypotheses to be tested through such designs Roberts, V., Sojo, V., & Grant, F (2020).

Natural experiments may arise when governing bodies introduce independent reporting, change selection procedures, or require safeguarding officers. Difference-in-differences designs can be informative where credible comparison groups and comparable pre-intervention trends exist. Interrupted time series can examine changes in multiple outcomes after a clearly timed policy change. These approaches must treat reporting as a distinct process: a better intervention may increase complaint counts even while the underlying incidence declines. Combining independent athlete surveys with administrative outcome data helps avoid mistaking changed detection for a treatment effect.

Mechanism testing should be ambitious but proportionate. A proposed mediation chain might run from external oversight through perceived credibility of reporting to earlier disclosure and more effective protective action. Each link requires measurement in a temporal order consistent with the proposed mechanism. Structural equation models do not by themselves overcome unmeasured confounding or reverse causality. Research should preregister plausible competing pathways, report uncertainty, assess heterogeneous effects, and openly publish analysis code or deidentified data when ethical and legal constraints permit. The ethical sensitivity of safeguarding datasets calls for privacy-preserving access rather than indiscriminate raw-data release.

Qualitative Methods and the Meaning of Organizational Experience

Qualitative inquiry provides access to processes hidden by incident counts. Athletes can explain why a particular coaching interaction was threatening, what they expected from a disclosure, and how peers or managers reacted. Parent's interviews with 27 sport stakeholders illuminate disclosure as a socially embedded process Parent, S (2011). Stirling and Kerr's interviews with retired swimmers generated behaviorally interpretable categories of emotional abuse Stirling, A. E., & Kerr, G. A (2008). Johansson's interviews and document analysis revealed how Swedish sports federations could formally oppose sexual abuse while having relatively few practical measures Johansson, S (2022). These designs generate explanatory mechanisms and reveal previously undermeasured dimensions.

Credibility in qualitative safeguarding research depends on appropriate sampling, transparent analytic decisions, reflexive attention to institutional position, and faithful preservation of participants' perspectives. Interviewing only organizational leaders will likely produce a different account from interviewing athletes who withdrew after a complaint. Participation may require the researcher to explain mandatory reporting obligations, confidentiality limits, and how interview data will be separated from ongoing disciplinary processes. The researcher must avoid inducing unnecessary traumatic retelling where the scientific question can be answered through less burdensome methods. These are methodological as well as ethical requirements.

Mixed-methods designs can connect reported mechanisms to broader patterns. For example, interviews may identify fear of scholarship loss; a survey may then estimate how frequently such fear is reported across programs, and administrative data may establish whether selection decisions changed after disclosures. A document review can test whether rules were actually applied when a well-connected staff member was involved. Such triangulation does not guarantee causal identification, but it makes claims more concrete and helps distinguish common organizational pathways from salient yet exceptional cases. The aim is analytic specificity, not numerical imitation of qualitative meaning.

Cross-National Governance and Cultural Transfer

Sport governance differs across national legal systems, educational structures, funding arrangements, athlete employment status, and child-protection responsibilities. International guidance supplies shared principles, while the practical implementation of protection and reporting must respect local law and institutional capabilities. The 2024 IOC evidence review found that published epidemiological studies were unevenly distributed across regions, with considerably more research from Europe and North America than from many other regions Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). That distribution limits claims about universal prevalence, institution types, and implementation pathways. Global models require evidence that includes diverse sporting systems and languages.

Cross-national transfer should focus first on core functions: recognizing prohibited conduct, protecting vulnerable individuals, independent access to assistance, impartial review, and organizational accountability. The form of a reporting office or its relationship to public authorities may vary. A practice effective in a well-funded national federation may be unrealistic for a volunteer club without adaptation. Rhind and colleagues' international safeguarding work highlights cultural sensitivity, resources, engagement, and networks as important implementation pillars Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F (2017). The framework points to dimensions to evaluate rather than providing a fixed administrative blueprint.

Comparative research must control for differences in definitions and reporting incentives. More recorded complaints in one jurisdiction could reflect more violence, broader behavioral definitions, stronger mandatory reporting, or a more trusted reporting channel. Researchers need matched instruments, information about legal requirements, and careful translation of conceptual terms. A global research program should recruit partners from underrepresented regions and ensure that athletes participate in interpretation, rather than using an externally derived category system without local validation. International comparability improves when measures retain stable core concepts alongside documented, justified contextual differences.

Ethics of Research, Confidentiality, and Participant Protection

Scientific investigation of safeguarding can itself create risks if an athlete becomes identifiable, a coach learns about a disclosure, or sensitive records are transferred across institutions without lawful protection. Research protocols should separate scientific data from disciplinary investigation where appropriate, specify foreseeable disclosures that may require referral, and inform participants about the limits of confidentiality. Special care is needed with minors, vulnerable adults, rare abuse events, and small national teams. Ethics review and informed consent are fundamental; potential participants should know what data will be collected, who can access them, and which protections apply during and after the study.

Data minimization can improve both ethics and measurement quality. Survey instruments need behaviorally specific items, but routine collection of names, identifying locations, or unrelated health information may increase danger without improving inference. Interviews can use pseudonyms and carefully edited contextual detail while preserving the meaning of experiences. Researchers should avoid promising absolute anonymity where repeated measures, small samples, or legal obligations make it impossible. Where independent data repositories are used, technical and contractual controls must be appropriate to the sensitivity of the material. These procedures are particularly important when a study is conducted with funds or access controlled by the institutions being evaluated.

Safeguarding research should document researcher independence, sponsor influence, conflicts of interest, and the protocol for communicating findings. An institution may resist the publication of harmful results or ask to identify participants whose reports suggest misconduct. Agreements made before recruitment should preserve researchers' ability to publish valid aggregate findings and participants' protection. Ethical responsibility also extends to how prevalence statistics and qualitative accounts are communicated: sensational or universalizing claims can distort evidence and create unnecessary harm. The aim is accurate public knowledge that improves prevention and accountability without treating survivors merely as sources of dramatic illustration.

Digital Communication, Surveillance, and Contemporary Risk

Sporting relationships increasingly extend through messaging, video analysis platforms, training apps, social networks, and remote coaching. These tools can support education, documentation, and access to independent services, while also enabling private approaches, unwanted contact, surveillance, and the circulation of sensitive images or complaints. The organizational mechanisms are familiar: authority, access, dependency, visibility, and trust. Digital channels change their speed, persistence, and reach. Research should examine whether boundaries established for in-person training remain effective in private messaging, temporary group chats, and international travel arrangements Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Administrative technology does not make a reporting system independent. An online form may route reports directly to the institution whose staff are implicated; an automated case-management tool may expose confidential information to administrators with unrelated responsibilities. Safeguarding requires transparent permissions, secure audit trails, retention limits, accessible alternative channels, and independent escalation. Automated triage may support organizational workflow, but any claim that an algorithm can classify allegations accurately or fairly requires representative evaluation, explicit error analysis, human oversight, and due-process safeguards. Sensitive disclosures should never be treated as ordinary low-risk administrative text.

The Artificial Era adds a research horizon in which organizations use artificial systems for communication, monitoring, and procedural assistance. Such systems may alter who receives information, how decisions are explained, and where control resides. The appropriate inquiry remains empirically grounded: can a tool protect privacy, preserve evidence integrity, reduce response delay, or instead increase misclassification and dependence? Sport safeguarding provides a concrete case for studying human–artificial governance relations, while the evidence presented in this dossier primarily concerns human institutions. Proposed artificial-system interventions are research hypotheses to be evaluated, not proven extensions of the established safeguarding evidence.

Research Gaps, Replication, and Program Priorities

The field has a strong foundation for documenting that multiple forms of interpersonal violence occur in sporting contexts. It has weaker foundations for estimating stable, internationally transportable prevalence under identical definitions, assigning causal weights to organizational risk factors, and identifying the most effective prevention programs Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024) Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J (2026). An immediate priority is measurement harmonization across age groups, sports, and countries while retaining context-specific questions. Another is the recruitment of people who left sport or declined formal disclosure. The study population must include the full trajectory of participation if the costs of harmful culture are to be understood.

A second priority is independent intervention evaluation. Federations could collaborate on preregistered, multisite studies of transparent selection rules, conflict-of-interest controls, protected reporting, supervised high-risk settings, and athlete participation in governance. Designs should track exposure, disclosure, retaliation, protective action, and well-being over time. Trials need ethically adequate baseline safeguarding for every comparison group; incremental interventions can be randomized or phased where safe and legally appropriate. A finding that awareness improved is insufficient to conclude that violence declined. Studies should register meaningful primary endpoints and assess sustained implementation under real competitive pressure.

A third priority is organizational mechanism research. Multilevel studies should examine when winning incentives affect tolerance of misconduct, how team norms shape peer behavior, whether confidential reporting trust changes disclosure, and how independent response affects perceived institutional betrayal. The theoretical literature offers pathways Roberts, V., Sojo, V., & Grant, F (2020) Morrison, E. W., & Milliken, F. J (2000) Smith, C. P., & Freyd, J. J (2014), and sport research offers concrete settings and observations Parent, S (2011) Johansson, S (2022). The missing link is often an adequately controlled temporal design. Independent replication, documented null results, and transparent disclosure of funding and institutional access are essential to prevent the research itself from reproducing reputational dependence.

Critical Synthesis of Competing Explanations

Several explanatory traditions illuminate safeguarding risk. A behavioral account focuses on harmful acts, reinforcement, and deterrents. A relational-power account examines dependency and the capacity to refuse. An organizational-culture account investigates normalized conduct and shared expectations. A governance account identifies conflicts of interest, incentives, and accountability. A socioecological account connects these mechanisms across levels. These approaches are compatible in many cases, but they generate different predictions. A culture-only intervention predicts improvement after norms change; a structural-power account predicts continuing danger where athletes remain unable to report independently despite improved messaging Roberts, V., Sojo, V., & Grant, F (2020) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024).

Evidence should discriminate among such explanations. If a federation introduces new values training but retains opaque selection and manager-controlled investigations, unchanged disclosure may be consistent with the importance of structural dependence. If external reporting becomes accessible but retaliation persists through informal team sanction, procedural independence alone will be insufficient. A design that measures culture, power, and institutional response separately can distinguish mediating pathways. Overly broad definitions of safe culture risk becoming unfalsifiable; an operational account names the proposed mechanism, identifies observables, and specifies which outcome would challenge the explanation.

The literature also contains tensions about the appropriate status of professional consensus. The IOC and FEPSAC documents organize complex findings and provide practical recommendations Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024) Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Mi (2025). They are indispensable to contemporary policy discussion, but they do not imply that all recommendations have the same empirical support. Similarly, cross-domain organizational work on psychological safety, silence, and betrayal supplies conceptual tools Edmondson, A (1999) Morrison, E. W., & Milliken, F. J (2000) Smith, C. P., & Freyd, J. J (2013). Its applicability to sport should be evaluated with local evidence, particularly because the sanctions for challenging authority in elite development may be unusually concentrated.

Evidence-Informed Institutional Design

A coherent safeguarding system can be organized as an accountable sequence. Risk recognition begins with specific definitions and mapping of high-risk relationships. Prevention addresses known opportunities for coercion, unmonitored contact, and conflicts of interest. Disclosure systems provide accessible options outside ordinary selection authority. Response mechanisms supply proportionate interim protection, trained investigation, communication, and appropriate external referral. Recovery provides sustained attention to the affected person's dignity, choices, and sporting future. Governance reviews the performance of every stage and corrects failures. This is a synthesis of functions identified across the IOC consensus and sport-specific organizational research Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024) Roberts, V., Sojo, V., & Grant, F (2020).

The sequence should be treated as a model to test. Its mechanisms may be disrupted by absent resources, social retaliation, poor role clarity, inappropriate information sharing, or leadership incentives. Evaluation must therefore identify not only how often reports arrive but also whether athletes were able to approach the system, whether concerns were addressed promptly, and whether harmful behavior or institutional betrayal was reduced. Administrative process indicators can be collected regularly, while confidential participant surveys and independent audits provide evidence about actual trust. The CHILDREN framework identifies culture, leadership, resources, engagement, and networks as related dimensions of system capacity Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F (2017).

Scientific integrity is maintained by giving each claim the weight of its underlying design. Retrospective prevalence research establishes experience under specified measurement conditions; qualitative interviews illuminate processes; systematic reviews identify patterns and limitations; consensus statements formulate shared guidance; controlled interventions can identify causal effects if well designed. Institutions should use all these forms of evidence without pretending that one replaces the others. The final practical standard is an organization capable of detecting harm, responding fairly, learning from credible evidence, and maintaining protection even when disclosure challenges celebrated personnel.

Conclusion

Safeguarding in sport is the organized relationship among protection, authority, culture, disclosure, response, and accountability. Interpersonal violence occurs within sporting environments shaped by unequal access to opportunities, intense performance incentives, and social norms that determine whether harmful behavior can be named and challenged. Historical investigations of sexual harassment, qualitative research on emotional abuse, large retrospective athlete surveys, organizational studies, and contemporary consensus statements converge on the significance of institutional context Fasting, K., Brackenridge, C., & Sundgot-Borgen, J (2003) Stirling, A. E., & Kerr, G. A (2008) Vertommen, T., Schipper-van Veldhoven, N., Wouters, K., Kamp (2016) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). The research object is therefore a system of conditions and practices whose operation can be investigated at athlete, team, organization, and wider societal levels.

The critical scientific result is a distinction between the evidence describing harm and the evidence demonstrating effective prevention. The literature documents abuse, dependence, silence, and policy-implementation gaps, while intervention studies capable of assigning causal effects to specific safeguards remain comparatively limited Roberts, V., Sojo, V., & Grant, F (2020) Johansson, S (2022) Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M (2024). Sound protection needs independent reporting and response structures, meaningful athlete voice, transparent decision rules, adequately resourced governance, and an ability to measure outcomes under competitive pressure. These are evidence-informed organizational priorities whose actual effects should be assessed through valid measurement and independent designs.

Future progress will depend on multisite longitudinal research, harmonized behaviorally specific exposure measures, inclusion of athletes who leave sport, controlled evaluation of prevention packages, and careful assessment of procedural fairness and institutional betrayal. The relevance of safeguarding extends to how artificial systems may mediate institutional communication and governance, but such extensions require their own empirical validation. The governing research principle remains stable: the safety of a sport system is demonstrated by its response to real vulnerability, the freedom of participants to speak and participate, and the documented correction of harmful institutional practices.

References

  • Brackenridge, C. H. (2000). Harassment, sexual abuse, and safety of the female athlete. Clinics in Sports Medicine, 19(2), 187–198. https://doi.org/10.1016/S0278-5919(05)70198-8
  • Carmo, E., Moreira, B., Azevedo, V., & Carvalho, J. (2026). Sexual violence in sport and a dual harm hypothesis of victimization and disrupted belonging: A systematic review. Psychology of Sport and Exercise, 88, 103278. https://doi.org/10.1016/j.psychsport.2026.103278
  • Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999
  • Fasting, K., Brackenridge, C., & Sundgot-Borgen, J. (2003). Experiences of sexual harassment and abuse among Norwegian elite female athletes and nonathletes. Research Quarterly for Exercise and Sport, 74(1), 84–97. https://doi.org/10.1080/02701367.2003.10609067
  • Johansson, S. (2022). From policy to practice: Measures against sexual abuse by Swedish sports federations. Frontiers in Sports and Active Living, 4, 841653. https://doi.org/10.3389/fspor.2022.841653
  • Khomutova, A., Chroni, S. A., Kavanagh, E., Ruffault, A., Miles, A., Moesch, K., Fontanesi, L., Nery, M., & Vertommen, T. (2025). FEPSAC position statement on safeguarding athletes in sport. Psychology of Sport and Exercise, 80, 102897. https://doi.org/10.1016/j.psychsport.2025.102897
  • Morrison, E. W., & Milliken, F. J. (2000). Organizational silence: A barrier to change and development in a pluralistic world. Academy of Management Review, 25(4), 706–725. https://doi.org/10.5465/amr.2000.3707697
  • Mountjoy, M., Brackenridge, C., Arrington, M., Blauwet, C., Carska-Sheppard, A., Fasting, K., Kirby, S., Leahy, T., Marks, S., Martin, K., Starr, K., Tiivas, A., & Budgett, R. (2016). International Olympic Committee consensus statement: Harassment and abuse (non-accidental violence) in sport. British Journal of Sports Medicine, 50(17), 1019–1029. https://doi.org/10.1136/bjsports-2016-096121
  • Parent, S. (2011). Disclosure of sexual abuse in sport organizations: A case study. Journal of Child Sexual Abuse, 20(3), 322–337. https://doi.org/10.1080/10538712.2011.573459
  • Rhind, D. J. A., Kay, T., Hills, L., & Owusu-Sekyere, F. (2017). Building a system to safeguard children in sport: The eight CHILDREN pillars. Journal of Sport and Social Issues, 41(2), 151–171. https://doi.org/10.1177/0193723517696966
  • Roberts, V., Sojo, V., & Grant, F. (2020). Organisational factors and non-accidental violence in sport: A systematic review. Sport Management Review, 23(1), 8–27. https://doi.org/10.1016/j.smr.2019.03.001
  • Smith, C. P., & Freyd, J. J. (2013). Dangerous safe havens: Institutional betrayal exacerbates sexual trauma. Journal of Traumatic Stress, 26(1), 119–124. https://doi.org/10.1002/jts.21778
  • Smith, C. P., & Freyd, J. J. (2014). Institutional betrayal. American Psychologist, 69(6), 575–587. https://doi.org/10.1037/a0037564
  • Stirling, A. E., & Kerr, G. A. (2008). Defining and categorizing emotional abuse in sport. European Journal of Sport Science, 8(4), 173–181. https://doi.org/10.1080/17461390802086281
  • Stirling, A. E., & Kerr, G. A. (2009). Abused athletes' perceptions of the coach–athlete relationship. Sport in Society, 12(2), 227–239. https://doi.org/10.1080/17430430802591019
  • Tuakli-Wosornu, Y. A., Burrows, K., Fasting, K., Hartill, M., Hodge, K., Kaufman, K., Kavanagh, E., Kirby, S. L., MacLeod, J. G., Mountjoy, M., Parent, S., Tak, M., Vertommen, T., & Rhind, D. J. A. (2024). IOC consensus statement: Interpersonal violence and safeguarding in sport. British Journal of Sports Medicine, 58(22), 1322–1344. https://doi.org/10.1136/bjsports-2024-108766
  • Vertommen, T., Schipper-van Veldhoven, N., Wouters, K., Kampen, J. K., Brackenridge, C. H., Rhind, D. J. A., Neels, K., & Van Den Eede, F. (2016). Interpersonal violence against children in sport in the Netherlands and Belgium. Child Abuse & Neglect, 51, 223–236. https://doi.org/10.1016/j.chiabu.2015.10.006
  • Willson, E., Kerr, G., Battaglia, A., & Stirling, A. (2022). Listening to athletes' voices: National team athletes' perspectives on advancing safe sport in Canada. Frontiers in Sports and Active Living, 4, 840221. https://doi.org/10.3389/fspor.2022.840221
  • Willson, E., Kerr, G., Stirling, A., & Buono, S. (2022). Prevalence of maltreatment among Canadian national team athletes. Journal of Interpersonal Violence, 37(21–22), NP19857–NP19879. https://doi.org/10.1177/08862605211045096