How do representatives from sporting organisations understand primary prevention of violence against women?

Sports settings have been identified as important locations for the prevention of violence against women, with numerous prevention initiatives currently running in many sports internationally. However, little is known about how those involved in sporting organisations, who are often tasked with delivering such initiatives, conceptualise the prevention of violence against women. This research draws on a survey of people who were invited to participate if they had professional experience in the development or delivery of violence prevention programs in their sporting organisation. We found that a cohort of participants had a limited understanding of primary prevention and how it applies to the prevention of violence against women through sport. Broadly, they were not aware of the difference between primary, secondary and tertiary prevention techniques. These findings suggest that there is a need for better education of those working in sporting organisations about the nature of primary prevention of violence against women as well as deeper consideration of the complex nature of doing violence prevention work through sport.


Introduction
Violence against women is a widespread concern that has deep negative impacts across the globe, with the World Health Organization (WHO) estimating that almost one in three women globally (30%) have been subject to sexual and/or physical intimate partner violence or non-partner sexual violence (WHO 2021). The United Nations Declaration on the Elimination of Violence Against Women defines violence against women as: any act of gender-based violence that results in, or is likely to result in, physical, sexual, or mental harm or suffering to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether occurring in public or in private life. (United Nations 1993) The term encompasses a broad range of harms that extend beyond physical and sexual violence to also include emotional, psychological, cultural, financial and technology-facilitated abuse. 1 Evidence-based strategies to prevent violence against women before it occurs are recognised as crucial by academics, practitioners and policy makers (Ellsberg et al. 2015;García-Moreno et al. 2014;Heise 1998Heise , 2011Michau et al. 2015;WHO 2002;Walden and Wall 2014;Our Watch 2021). The most influential approach is based on a public health framework, which allows for the development of prevention strategies at three different points of intervention: primary, secondary and tertiary. Primary prevention aims to move upstream to address the root causes or drivers of violence, whereas secondary prevention targets at risk groups and tertiary prevention attempts to reduce the risk of recurrence (WHO 2002). Identifying both the context and drivers of violence against women allows for the identification of multiple complex and reinforcing measures required to effect positive change and their implementation in the various settings where people live, work and play (Walden and Wall 2014;Wilmerding et al. 2018).
A primary prevention approach relies on a social-ecological model which takes into consideration the multiple protective and risk factors that might put an individual at lesser or greater likelihood of committing violence against women (Wall 2014). These factors play out across the various levels of the social ecology: from individual and relationships, to community, economy, culture and society (WHO 2002). Examples of risk and protective factors can include personal characteristics, peer attitudes, family experiences, community norms and broader societal or structural influences (Heise 1998(Heise , 2011.
One recognised setting where people can both learn and reproduce particular attitudes, behaviours and social norms is sport. Alongside school and work, sport is a key institution of socialisation where people engage with and learn from their peers. Furthermore, sportspeople appear frequently in mass and social media and are widely considered to be influential as role models to children and young people.
Research has found that male-dominated peer and professional networks are significantly more likely to hold attitudes supportive of violence against women and gender inequality (Webster et al. 2018;Durán et al. 2018;Seabrook et al. 2018). Conversely, men with a diverse range of social networks are less likely to hold violence supportive attitudes (Kaczkowski et al. 2017). Male-dominated sports may promote limited and stereotypical forms of masculinity that foster and even endorse violent behaviours (Albury et al. 2011;McCauley et al. 2014;Ralph and Roberts 2019;Sønderlund et al. 2014). Collective norms of masculinity can impact men's resistance to change (Burrell 2021;Stewart et al. 2021). Moreover, "cultures of impunity" can exist in sporting teams where violent supportive attitudes and behaviours can go unpunished and even celebrated (MenEngage and UN Women 2015, pp. 31-32;Corboz et al. 2016). As a result, some male-dominated peer contexts and organisational cultures can be intimidating and dangerous for women who come into contact with them or their members, in relationships, socially, through family or as players (Breger et al. 2019;DeKeseredy and Schwartz 2013;Schwartz 2021).
The need to engage men and boys in the prevention of violence against women has been widely acknowledged as an important strategy and in recent decades there have been increasing attempts to do so (Casey et al. 2018;Glinski et al. 2018;Gibbs et al. 2015;Keddie and Bartel 2021;Kimball et al. 2013;McCook 2022;Pease 2019). Interventions with men and boys at the group and community level have been noted as a key setting to implement strategies to improve collectively-reinforced norms around the acceptability of violence against women (García-Moreno et al. 2014;Stewart et al 2021). There is increasing evidence that success in engaging men and boys lies not just in focusing on changing individual violence-supportive attitudes and propensity for violent behaviour, but rather enacting approaches that, in line with the aforementioned public health model, transform identities, relations, policies, practices and cultures in such settings (Carson et al. 2015;De Gue et al. 2012;Dickson and Willis 2017;Our Watch 2019;Our Watch 2021). The most effective prevention work appears to address gender inequality across multiple levels of target organisations and incorporate multiple interrelated components rather than just one-off activities (Nation et al. 2000;Allen et al. 2019;Glinski et al. 2018;Stewart et al. 2021).

Previous studies
A number of studies have examined how primary prevention is conceptualised and articulated by organisations that engage in it, along with how those organisations then operationalise their prevention activities. For example, a 2014 commissioned review of Australian service providers delivering primary prevention and early intervention programs with relevance to men and boys found that practitioners, some of whom were involved in sport, had a good understanding of the ecological model of prevention (Carmody et al. 2014). The 37 participants were able to articulate that violence against women is caused by inequality and violence supportive norms. Despite this, the researchers found that prevention practice involving men and boys was in its nascence and that the activities undertaken at the time were "piecemeal, ad hoc and dispersed" (p. 70). The researchers also found a paucity of actions that could be considered primary prevention. Their recommendations included better and more comprehensive training for those working with men and boys along with an increased focus on genuine primary prevention.
Limited heterogeneity in delivery methods was echoed by a New Zealand study which surveyed 44 respondents from 42 agencies engaged in activities described as the primary prevention of sexual violence (Dickson and Willis 2017). The researchers found that the bulk of their efforts focused on education and awareness-raising. They concluded that this may be a result of a low level of knowledge about, and a lack of sustained commitment to, primary prevention more broadly (Dickson and Willis 2017, p. 140). Similarly, awareness raising (of themselves and others) along with attendance at events, were the most commonly reported activities in a survey of 379 men who were involved in gender-based violence prevention across the globe .
Lack of distinction between primary, secondary and tertiary prevention was highlighted in a study of international anti-violence organisations that engage men in prevention. Storer et al. (2016) conducted 29 interviews with frontline service providers and found that the participants did not neatly delineate between primary, secondary and tertiary prevention. Most interviewees had an opaque interpretation of the nature of primary prevention, and were often focused on actions rather than outcomes. The research identified gaps between how violence prevention is framed and how it is actually delivered-a product, they suggest, of the many competing demands placed on those at the front lines of violence prevention. The findings led the researchers to question the relevance of rigid Western public health frameworks and to propose locally and culturally relevant training underpinned by theory (Storer et al. 2016, p. 265; see also Pease 2019).
The gap between theory and delivery was also identified in interviews with 16 stakeholders specifically involved with the implementation of prevention programs in sport settings in Australia and New Zealand. The researchers found that those working on the front line of prevention delivery rarely categorised their work as primary, secondary or tertiary prevention (Hamilton et al. 2020). Instead, participants mostly spoke of primary prevention in broad terms of culture change. The research also found that some sporting organisations, for example some regional clubs, did not feel that they or their community were ready to engage with prevention explicitly. This research raises questions about the level of understanding of the nature of primary prevention and again highlights some of the barriers that prevent such a model being widely understood and adopted.

Rationale for study
Effective programs require implementation by people who have a strong understanding of what prevention actually entails. However, the literature suggests that there are gaps between how prevention is conceptualised and how it is implemented. To that end, it is important to gauge how sporting organisations understand and apply prevention principles in order to ensure that actions are in line with any policy goals of primary, secondary and tertiary prevention, to identify and clarify any misunderstandings and to maximise sporting organisations' potential for effecting long-term changes in the prevention of violence against women.

Methodology
The study received ethics approval from RMIT University before it commenced. The research was conducted between August and December 2018.

Participants
In seeking participants, we purposefully targeted sporting codes such as Australian cricket, soccer, and basketball (including wheelchair basketball) for three main reasons. Firstly, such sporting codes have national and sub-national leagues for both women and men, include professional and non-professional leagues. Along with being amongst the most played team sports for men and boys (Australian Sports Commission 2016) they have a wide audience and support base for national and subnational teams. These characteristics make these codes ideal for examining understandings of preventing violence against women, given the potentially broad reach and impact of any primary prevention efforts focused on working with men. Secondly, previous research on violence prevention and sport in Australia has tended to focus on the popular spectator sports of Australian rules football and rugby league (e.g., Carmody et al. 2014;Corboz et al. 2016;Dyson et al. 2010;Albury et al. 2011;Dyson and Corboz 2016). Little published research appears to examine how other popular team-based sports understand or implement the prevention of violence against women. Thirdly, sports such as cricket, soccer and basketball are played internationally, therefore, the findings may offer insight into sporting organisations in other jurisdictions.
After conducting high level mapping of the national and state organisation of each sport, participants were contacted via publicly available email addresses on sporting organisation websites in Australia. Participants were invited to complete the survey if they were someone with broad professional experience in the development and/or delivery of primary prevention approaches in a sports setting, including increasing participation by women and girls, community development/engagement and diversity. These could be prevention and inclusion programs more broadly and were not restricted to those targeting violence against women. Snowball sampling was also utilised whereby recruited participants were invited to share the survey link to other relevant and interested parties.
A total of 114 email invitations were sent to potential participants, with the subject heading: "Invitation to complete online survey about the prevention of violence against women and sporting organisations." After a two-month period, there were 46 recorded entries in Qualtrics at the close of the survey. Following data cleaning, the final data set comprised 29 complete responses, with one removed due to lack of consent and 16 removed from lack of response to substantive questions (i.e., responded only to initial questions on sport and/or role in organisation). This represents a final response rate of 25%, which is comparable to other survey research with representatives from organisations (Baruch and Holtom 2008). The final sample included participants with roles in sporting organisations such as coaches, general managers, club development managers, senior executives, female participation officers and athletes. Participants will hereafter be referred to as respondents or 'representatives' of sporting organisations.

Survey design and implementation
In order to address the aims, our research design adopts a qualitative methodology, involving an online survey with stakeholders from select sporting organisations. Qualitative research draws on an interpretivist epistemology in that it stresses an "understanding of the social world through an examination of the interpretation of that world by its participants" (Bryman 2004, p. 266).
A survey with open-ended questions enables interviewees to describe their experiences and knowledge of sporting programs and prevention work in their own words.
The online stakeholder survey was developed and distributed using the platform Qualtrics. The survey collected brief information about respondents' professional role(s) and associated sport before moving to three open-ended questions that examined representatives' perceptions of the actions and activities necessary for the primary prevention of violence against women. The survey also asked representatives to describe any internal or external work (if any) that their sporting organisation engages in to prevent violence against women. Finally, the survey included a 5-point Likert scale to measure representatives' perceptions of the importance of eight key elements of primary prevention for violence against women: (i) changing norms and cultures; (ii) education and awareness-raising; (iii) promoting gender equality; (iv) challenging rigid gender roles and stereotypes; (v) working across the whole of population (vi) challenging the condoning of violence against women; (vii) promoting women's independence and decision making; and (viii) strengthening positive, equal, and respectful relationships.
The elements of the survey were drawn from national and international frameworks for the primary prevention of violence against women (e.g., Our Watch 2021; WHO and London School of Hygiene and Tropical Medicine (LSHTM) 2010). Such an approach has been adopted in similar previous research regarding the conceptualisation and application of primary prevention of violence against women work by global anti-violence organisations working with men (Storer et al. 2016). We have also drawn on previous research with sporting organisations (Liston et al. 2017) and the work of Storer and colleagues (2016) to develop survey questions that will directly and indirectly tap into sporting organisations understandings of prevention work.

Data management and analysis
Basic descriptive analysis was used to analyse representatives' responses to the Likert scale question. A combination of qualitative conventional content analysis and summative content analysis were used to organise and interpret the data from the three open-ended questions (see Hsieh and Shannon 2005 for an overview). Survey responses were initially coded in Microsoft Excel, and were further categorised and refined using NVivo 12. All themes were latent and inductively driven by the data. Subthemes were quantified by counting certain words (e.g., education), with the aim of understanding how the usage of those words conveyed representatives' perceptions and conceptualisations of primary prevention of violence against women. Themes and subthemes were validated through independent cross-checking by each member of the research team.

Results
The online survey took between approximately 2 and 24 min for representatives from sporting organisations to complete (M = 7.82, SD = 6.40). The length of their responses to the 3 open-ended questions also varied widely between 1 and 237 words (M = 16.07, SD = 30.62). The qualitative analysis of these responses indicated mixed understandings of primary prevention of violence against women. The main findings are discussed below under the following two headings: (i) limited understanding of and engagement with primary prevention; and (ii) basic to strong understandings of primary prevention.

Limited understanding of and engagement with primary prevention
The online responses indicated that some representatives from sporting organisations lacked a clear understanding of primary prevention in general, and particularly how it applies to the prevention of violence against women. This was indicated by either (i) responses that were not related to gender or violence against women, or (ii) responses that were examples of secondary or tertiary prevention. For example, one respondent commented on the importance of ensuring refugees were active regardless of background and culture, while other responses appeared to focus on issues such as child protection, LGBTQI rights and broader psychological concepts such as jealousy and greed. In total, 15 responses referred to issues outside the realm of gendered violence, or made very unclear connections to the prevention of violence against women.
Other responses were connected to gendered violence, but did not demonstrate a clear understanding of how the actions were linked to preventing violence before it occurs. Eight responses described actions that were better categorised as secondary or tertiary prevention. For example, one respondent commented on domestic violence leave that was available and promoted to staff who had directly experienced violence by a family member. Several other respondents referred to actions such as hiring security staff for sporting events, and ensuring "severe punishments and condemnation for perpetrators." Such actions may be better conceptualised as early interventions or tertiary responses as they target those at risk of experiencing or perpetrating violence, or focus on preventing violence from reoccurring.
Survey responses also indicated that many of the sporting organisation representatives did not engage in primary prevention of violence against women. Twelve respondents stated that their sporting organisation did no or very little external work to prevent violence against women (e.g., work with the community); three respondents said their organisation did not engage in any internal violence prevention work (e.g., policies, in-house training); and eleven respondents were uncertain or unaware of any prevention work undertaken by their organisation. One representative, however, stated that while their sporting club was not engaged in any specific work to prevent violence against women, "individually there [were] people involved at different levels." Seven respondents also indicated a lack of awareness of any expert individuals or organisations to consult for information about violence against women.

Basic to strong understandings of primary prevention
Survey responses also indicated other representatives of sporting organisations appeared to have more established understandings of primary prevention and how it applies to violence against women. Some representatives appeared to have relatively surface level understandings of primary prevention of violence against women, while others appeared to have more comprehensive understandings. The responses also indicated that some actions to prevent violence against women are more widely and better understood than others. These actions will be discussed in turn.
The majority of representatives appeared to understand that education for members within a sporting organisation, as well as the broader community, was important for preventing violence against women before it occurs. In total, 35 responses referred to 'education and awareness raising' and 100% of participants perceived these actions to be moderately important (6.9%) or very important (93.10%). Many representatives briefly cited 'education' without elaboration, while others were somewhat more descriptive. For example: Education around 'banter' and how it leads to inequality that can then lead to violence against women. -Survey 21 (Committee member and Female Participation Officer) One representative stated that it was important to raise awareness that "violence against women comes in different forms" (Survey 24, Athlete). This comment appeared to be particularly relevant, as another representative seemed to emphasise physical violence over other forms: Education of men on respect for women; generation of boys knowing it's never ok to physically assault a woman. -Survey 7 (Operations Manager) When asked what actions and activities were necessary for the prevention of violence against women, 'attitude, behaviour and culture change' was another frequent response (n = 12). Changing norms and cultures was perceived as the most important primary prevention action across the sample (see Table 1), and 100% of participants perceived this action to be moderately important (3.45%) or very important (96.55%). Representatives also commonly emphasised the importance of changing attitudes, behaviours and cultures by targeting children and young people. Six responses referenced the importance of starting young, while eight referred to positive role-modelling for young people. For example: It [primary prevention] is having strong role models and instilling what role model behaviour is in our young children. Dealing with destructive behaviours in both genders early (i.e., in teenage and young adult years). -Survey 23 (Senior Executive) Workshops, codes of conduct introduced at a young age (with parent involvement) as to what is acceptable behaviour.' -Survey 26 (Coach) Raise boys to respect/value women and girls. -Survey 2 (Competitions Administrator) Education from parents and good role modelling from a very young age, reinforced by education and good role-modelling in schools, reinforced by media messaging. -Survey 18 (Diversity and Inclusion Specialist) It was clear from the survey responses that parental involvement with young people was thought to be key in changing attitudes and behaviours that are associated with violence against women. This appeared to be perceived as more important than the involvement of members from sporting clubs and organisations (e.g., coaches). Similarly, three representatives acknowledged the media's responsibility in changing cultures that condone violence against women. For example: Media seem to cover and take more importantly the consequences of misconduct on the field than violence against women. The outcry from the public is huge if an AFL (Australian Football League) player is out for a match or two Table 1 Responses to Likert survey question on personal perceptions of the importance of different key elements for the primary prevention of violence against women and girls Across all 8 elements, no respondents selected 'Unsure' and so this column has not been included here When asked what internal work (if any) their sporting organisation engaged in to prevent violence against women, some representatives mentioned policies that promote gender equality and women in sport. Five representatives referred to actions that promote women's leadership, representation and participation. For example: We are actively targeting staff gender balance and proactively fast-tracking high-potential women into leadership roles. -Survey 1 (Manager) Strong push for gender split on boards, push for more female coaches and female participation and identifying good and bad cultures across the state. -Survey 2 (Competitions Administrator) Despite these comments, the promotion of women's independence and decision making was the second least recognised action across the sample to prevent violence against women before it occurs (see Table 1). One representative rated this action as 'not important', while just under 69% of participants rated this action as 'very important.' Likewise, 'challenging rigid gender roles and stereotypes' was perceived as the least important primary prevention action, with the following ratings: not important (3.45%), important (6.9%), moderately important (27.59%), and very important (62.07%).

Discussion
Sport settings have been identified as important locations to engage men and boys in the prevention of violence against women. Yet the success and longevity of various prevention initiatives is often underpinned by the understanding, competency and commitment of trainers and implementers (De Gue et al. 2014;Mihalic et al. 2004). Our study sheds light on how representatives from sporting organisations understand and engage with the primary prevention of violence against women. In line with previous research (Hamilton et al. 2020;Storer et al. 2016), representatives rarely conceptualised primary prevention according to Western public health frameworks; they tended to cite a mixture of primary, secondary and tertiary actions to indicate their knowledge of how to address violence against women before it occurs. This suggests insufficient training about the nature of primary prevention, which in turn suggests a lack of organisational commitment. This may prevent sporting organisations from either engaging in those actions, or acknowledging and celebrating primary prevention that may already be occurring. Representatives also demonstrated varied understandings and perceptions of key elements known to be associated with the primary prevention of gendered violence. In general, representatives recognised that broad cultural and attitudinal change, public education and awareness raising were required to prevent violence against women. Some also emphasised the importance of targeting children and young people to intercept and create better attitudes and cultures in future generations. While potentially promising indicators of their knowledge, many representatives mentioned 'education' and 'awareness-raising' as necessary primary prevention strategies without further elaboration. Researchers and practitioners have pointed out that awareness-raising is often equated to prevention, and that while it is an integral element, community-based prevention initiatives often never progress beyond the awarenessraising phase, which in-and-of-itself does not lead to widespread behaviour change (Dickson and Willis 2017;Graffunder et al. 2011;Stewart et al. 2021;Storer et al 2016). The anti-violence activities undertaken within organisations are not necessarily delivered by committed and trained staff, nor are they sufficiently documented and evaluated (De Gue et al. 2014;Dickson and Willis 2017, p. 133). Graffunder and colleagues (2011) have highlighted that effective social change requires action beyond mere dissemination of information. They have also argued that strategies which primarily target young people will be insufficient if they do not also target the community and cultural systems that influence youth.
Representatives were also less inclined to emphasise the importance of other key elements. In particular, the promotion of women's independence and decision-making, and the challenging of rigid gender roles and stereotypes were viewed as the two least important primary prevention strategies. This finding is concerning given the wide body of research that has connected these actions to the sustained prevention of violence against women (WHO and LSHTM 2010;Our Watch 2021;Ellsberg et al. 2015;García-Moreno et al. 2014). It indicates a potential lack of engagement with feminist principles of equality and social justice-aspects which have been emphasised as integral in the effective implementation of prevention programs to end violence against women (Pease 2019). Furthermore, it suggests that these organisations may not be ready to engage in more complex 'gender transformative' prevention work where rigid, idealised and harmful forms of masculinity are challenged and altered (Casey et al. 2018;Gibbs et al. 2015;Our Watch 2019;McCook 2022).
These findings support the provision of more specific information and education about key drivers of violence and essential actions to prevent it may be useful to those working in sporting organisations (Hamilton et al. 2020). This would provide a foundation for engaging men and boys in more complex and multi-pronged strategies to transform identities, relations, policies, practices and cultures in such settings Carson et al. 2015;De Gue et al. 2012;Glinski et al. 2018;Our Watch 2021;Stewart et al. 2021).

Limitations
While our study provides insight into the knowledge and perceptions of representatives from Australian sporting organisations, it is important to recognise that such participants are not representative of all sports, and there may be response bias from those that chose to complete the survey. The online survey method restricted our ability to clarify responses and follow-up to get richer details about participants' understandings. Nevertheless, the brief responses from some participants may be indicative of their knowledge. As Ackerly and True (2008) point out, what people do not say or what is silent or absent in research can be just as revealing as what is said.
Even though the respondents were not necessarily implementers of gendered violence prevention programs, and likewise that not all contexts in which they work allow time and space for an expansive focus on primary prevention, a whole-of-sport framework (Our Watch 2020) suggests that it is still important that representatives are aware of the key elements and drivers of violence against women, and how best to respond to these. The last decade has seen significant attention paid to sport as a setting for the primary prevention of violence against women, as well as increasing commitment of sporting organisations in this endeavour. It would be fruitful for future research to track any progress in implementers' and representatives' understandings of prevention principles and initiatives. Our survey offers an efficient way to measure genuine responses in an anonymous format over time. Further interviews would add additional context and depth.

Conclusion
Responses from our online survey, paired with previous research (Hamilton et al. 2020), suggest that representatives from sporting organisations need greater education to improve their understanding of primary prevention of violence against women. Survey findings also revealed a need for broader organisational support for primary prevention initiatives so that the responsibility does not fall onto individuals and so that representatives recognise the responsibility of sporting clubs and organisations to enact prevention principles rather than deflecting accountability to external entities (e.g., parents, media). It was also evident that representatives require greater awareness about the forms of violence against women can take (e.g., beyond physical assault) and the individuals and organisations to consult for information about violence against women.
It is vital that implementers of prevention programs are aware of the complexities of educational content, delivery methods, and the broader range of primary prevention actions if initiatives are to be successful. Ensuring the knowledge and competency of representatives and implementers will enhance the potential of sporting organisations to engage the community, particularly men and boys, in the long-term prevention of violence against women.
Funding Open Access funding enabled and organized by CAUL and its Member Institutions.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission