Coach perceptions of child safeguarding policy in New Zealand’s youth sport sector

ABSTRACT

The current study surveyed 237 coaches in New Zealand on their perceptions of child safeguarding (CSG) discourse, resources, and policy in sport; and the changes in pedagogy they deemed necessary in response. Data indicates that, despite a reasonable level of awareness and uptake of CSG policy in the sector, concurrent trends require attention. Specifically, this relates to the potential adoption of defensive pedagogies as a result of unproblematic misinterpretations of CSG policy. We conclude with the suggestion that “critical” coach development learning contexts and experiences are needed to encourage coaches to discuss the application of CSG procedures in sport specific settings with experts; whilst also considering appropriate pedagogies dependant on athlete age, gender, and competitive levels. Importantly, actors in these spaces must consider whether (mis)interpretations of CSG are leading to the uptake of defensive pedagogies that may be unhelpful and detrimental to the development of children through sport.

KEYWORDS:

• Sport coaching
• child safeguarding
• policy
• sport
• defensive pedagogies

Introduction

Sport has been described as a possible location for the abuse and maltreatment of athletes (Brackenridge et al., 2012; Brackenridge, 2002). Indeed, increased media coverage (e.g. George, 2020; Hurley, 2019; Pesta, 2019) and an emerging body of empirical research indicates an ongoing need for vigilant consideration of the causes and effects of failing to safeguard young people in sport (Mountjoy et al., 2015). Recently, sport administrators have been spurred to consider the legal, moral, and civil consequences of restricting athletes’ opportunities to exercise their rights. In turn, calls have been made for more robust child safeguarding (CSG) efforts (e.g. Brackenridge et al., 2012; Human Rights Watch, 2020). Encouragingly, this has coincided with consideration of “child-focused” approaches to CSG policy – that is, a recognition of the importance of healthy and contributing citizens, the promotion of children’s equal opportunities, and overall athlete wellbeing (Lang & Hartill, 2014).

As Öhman and Quennerstedt (2017) observe, however, procedural documents can often include statements on what has been deemed by the institution to be appropriate and/or inappropriate interaction. Subsequently, coaches can often view these documents as misaligned with the realities of their coaching contexts. For instance, previous scholarship has highlighted the way(s) in which well-meaning CSG procedures are, in fact, casting the contemporary coach as both a predator in, and a protector of youth sport (Piper et al., 2013b; Taylor et al., 2016). As such, implementation of CSG policy has been reported as a major factor – for better or worse – in shaping coach pedagogies (Lang, 2010, 2015; Piper et al., 2013a, 2013b).

Mindful that CSG measures are undoubtedly implemented in the spirit of protecting the positive outcomes of intergenerational engagement – be it in sport or any other educational setting – the current study takes a lead from trends in the global literature that critiques the unproblematic adoption of defensive pedagogies resulting from misinterpreted CSG policy (Lang, 2010; Piper et al., 2012; Öhman & Quennerstedt, 2017). Indeed, the lived experiences of the first and third authors – that is, our observation of others’ and our own assumptions about CSG policy (see Bennett et al, 2022) – also suggested that more research was needed in the New Zealand context. Specifically, the aim of the current study was to examine New Zealand coaches’ understandings of CSG discourses, resources, and policy; and the types of pedagogical changes they perceive are necessary in response.

Literature review

Given the aim of this study, the behaviours that warrant improved safeguarding measures in sport need to be considered. Young (2012) suggests that sports related violence (that is, the maltreatment of athletes within the context of sport) encompasses direct acts of physical violence, either within or outside the sanctioned rules of the sport, that result in injury; and/or that which threatens, causes injury, or violates human justices and civil liberties. Parent and Fortier (2018) suggest that this definition is sufficiently broad to encompass both relational maltreatment of athletes, such as sexual, psychological, or physical abuse, and/or neglect; and non-relational maltreatment including (but not limited to) harassment, hazing, and violence between players during a competition (Stirling, 2009).

The breadth of this description is important given the equally expansive circumstances that athletes can be exposed to violent acts. For instance, research suggests that sport related violence can occur in team social events, in changing rooms, and during club/team initiations (Waldron, 2015), over social media (Evans et al., 2016), during competition (Gendron et al., 2011; Smith, 1983), or at the home of the coach (Brackenridge et al., 2010; Mountjoy et al., 2015). These settings have been known to harbour violent practices that range from humiliation, bullying, and physical assault (Collot D’Escury & Dudink, 2010), as well as sexual assault (Alexander et al., 2016; Elendu & Umeakuka, 2011; Fasting et al., 2011, 2014). Thus, the scholarship to date suggests that, globally, violence against athletes is prevalent, posing significant issues for those involved to acknowledge and address (Alexander et al, 2016; Evans et al., 2016; Fasting et al., 2014).

It has been argued that current sport systems do not adequately safeguard young athletes from violent and/or abusive practices (Brackenridge et al., 2010). For instance, cultures of obedience in sport (End Game 2020; Parent & Fortier, 2018) and an over-emphasis on success and performance are considered to be contributing causes to the occurrence of violence towards athletes (David, 2005). Thus, scholars maintain that the promotion of athlete health, safety, wellbeing, and rights must be emphasised over performance outcomes (Brackenridge et al., 2010; David, 2005; Donnelly, 1997, 2008; Lindhorst, 2015; Weber, 2009).

Encouragingly, CSG policy and resources have begun to emerge to guide sport sector stakeholders (i.e. coaches, athletes, parents, administrators, etc.) on “best practice” approaches, and address the reported prevalence of athlete maltreatment (e.g. Ministry of Education, 2014; Brackenridge et al., 2010; Mountjoy et al., 2015). Problematically, however, scholarship has also identified some noteworthy shifts in coach pedagogy as a result of misinterpretations of policy, and increased social discourse on the matter of athlete maltreatment. For instance, commenting on the state of sport coaching in the United Kingdom, Taylor et al. (2016) suggested that mainstream discourses about child safeguarding, the associated procedures such as police vetting, codes of conduct encouraging “no touch” approaches, and an overall sense of increased surveillance, have contributed to a culture of suspicion in sport coaching circles. Lang (2010) found that coaches often feel at risk of being accused by parents of wrongdoing, and therefore regulate their pedagogy in line with dominant notions of CSG (i.e. “no touch” pedagogies) more rigorously when parents are present. Piper et al. (2012) suggests that this anxiety is seated in a “moral panic” (Johnson, 2000), and promotes the perception that adults in coaching roles deserve disproportionately rigorous scrutiny.

Similar influences on coaching practices can also be observed in the United States. At the turn of the century, Dickinson (2000) cautioned coaches to consider the time, place, and context in which they make physical contact with their athletes. Dickinson’s suggestions implied that in the event of injury, a coach is well advised to carefully consider the necessity for physical contact if they are not accredited to do so; and that even the action of hugging an athlete may be considered a criminal offence if circumstances are misconstrued. More recently, the prerequisite of police vetting in the United States has been extended to include compulsory fingerprinting of all professional and volunteer adults engaged in children’s sport (for example see Howell Soccer Club, 2018). Piper et al. (2013a) problematise this trajectory by suggesting that: “if the way of keeping everyone safe is to create a situation in which children and adults each see the other as dangerous and toxic, then significant collateral damage to the quality of in loco parentis¹ relationships appear inevitable” (p. 578).

Child safeguarding in New Zealand

While Lang and Hartill’s (2014) work highlights the efforts of, and issues faced by many countries around the globe, it is the roll out of the “Children’s Act” (Ministry of Education, 2014) in New Zealand, and the subsequent CSG efforts in the New Zealand sport sector that form the basis for the current study.

The Children’s Act outlines a range of requirements intended to influence the practice of those working alongside children in various vocations. The foundational purpose of this legislation was to increase the probability that children will be safe/safeguarded when interacting with adults. For example, according to Section 9.2 of the Children’s Act (2014), this purpose includes: a) protecting children from abuse and neglect; b) improving their physical and mental health and their cultural and emotional well-being; c) improving their education and training and their participation in recreation and cultural activities; d) strengthening their connection to their families, whanau (family), hapū (descent group), and iwi (tribe), or other culturally recognised family group; e) increasing their participation in decision making about them, and their contribution to society; and f) improving their social and economic well-being. This legislation requires that child protection policies be adopted and reported on by prescribed State services, District Health Boards (DHBs), and school boards; and be adopted by those who enter into contracts or funding arrangements with said boards (Children’s Act, 2014, Section 2.14).

Clearly, the legal requirement to conceive and implement CSG policy has important implications for schools and teachers. However (and more in scope for the current study), there are distinct implications for Sport New Zealand (Sport NZ) – the Crown entity responsible for governing and funding sport in New Zealand, and that frequently works at the school-sport interface to promote and direct the “codes of ethics” for coaching and teaching (Bennett et al., 2022). In response to the Children’s Act, guidance and training resources released by Sport NZ have included a suite of 12 adaptable policies and procedures for community organisations (October, 2020); an online module for “Child Protection in Play, Active Recreation and Sport” (October, 2020); and eight “bite-sized” modules aimed at creating safe, positive environments for young people taking part in active recreation and sport (September 2021) (J. Richards, personal communication, 19 April 2022).

Sport NZ’s CSG-focused online courses emphasise the role of the coach/organisation to observe for signs of danger, by recognising, responding to, reporting and recording incidents and allegations of misconduct. Coaches, as adults in charge of young people, are positioned as people in roles of “trust” with a duty of care (i.e. to observe for, and report signs of danger/risk). The ways in which adults can and should rearrange their practice and behaviours is conveyed through a list of segments related to the “do’s and don’t’s” of touch, photography, one-on-one’s with children, using bathrooms/changing rooms, pick-ups, travel, and billets. As an example of the latter, it is suggested that when billeting young people as part of a sports trip, the visiting team should request to view two forms of ID of the host, carry out a risk analysis on the host family, and run a police check on anyone in the house who is over the age of 18 years (Sport NZ, 2022a).

Hypothesis

Research suggests that many CSG initiatives are predicated on deficit perspectives that prioritise a protection discourse over safeguarding discourse (Öhman & Quennerstedt, 2017). Fuelled by misinterpretations of legislation (Lang & Pinder, 2017), coaches/educators resort to defensive, hands-off approaches in the hope that their pedagogy will not be misconstrued as “dangerous” (Johnson, 2000), nor place themselves “in danger” of accusation (Taylor et al., 2016). Similarly, misinterpreted policies can result in (volunteer) coaches in a position to discern potential danger in the practice of others (Autor et al, xxx). Ironically, the unproblematic incorporation of this “trichotomy of danger” - initially intended to protect the vulnerable child – is simultaneously driven by notions of self-protection; and the idea that being defensive may reduce the presence of risk/danger (Lang, 2010, 2015; Piper et al., 2013b). As such, monitoring techniques such as police vetting and fingerprinting (see Howell Soccer Club, 2018) can act to amplify the fear of accusation among human movement educators (Johnson, 2013), leading to wholesale pedagogical adjustments that are neither advocated for, nor desirable (McWilliam & Sachs, 2004). It is critical to acknowledge, too, that defensive pedagogies can also limit a child’s right to participation in positive, tactile learning experiences (Gleaves & Lang, 2017; Piper et al., 2012; Öhman & Quennerstedt, 2017.

As the current study sought to understand how New Zealand coaches’ made sense of child CSG discourses in their own practice, and the types of pedagogical changes they perceive were necessary as a result, we hope to add to the discussion presented by Bennett et al., (2022) concerning the three-fold notion of danger coaches encounter in their work (i.e. “in danger”, “dangerous” and “observing for danger”). In the course of examining participants’ understandings of CSG discourses, resources, and policy, our research also sought to examine if/how coaches were reorganising their pedagogy in ways that, to their mind, allayed any risk of “danger” in practice.

Methods

Participants

The current study focused on New Zealand sport coaches. Participants were all over the age of 18, and were required to be coaching athletes, or had recently been coaching athletes, in any code, under the age of 18 (i.e. a “child”; Ministry of Education, 2014). Given Piper et al. (2012) observation about the possible issues faced by the volunteer coaching population (a segment that makes up a large proportion of New Zealand’s coaching “workforce”), it was initially intended that all participants should identify as a volunteer sport coach. Nevertheless, as a broader view of the sector was desirable, data from both volunteer and paid coaches was collected. This was later extrapolated and compared in the data analysis phase. Demographic data of the participants was requested (see Table 1) in order to reflect the diverse, and therefore complex makeup of New Zealand’s coaching workforce.

Table 1. Demographics of the participants (N = 237).

	Counts	Percentage
Coach age
18–20	12	5.1 %
21–25	21	8.9 %
26–30	9	3.8 %
31–35	9	3.8 %
36–40	26	11.0 %
41–45	47	19.8 %
46–50	41	17.3 %
51–55	32	13.5 %
56–60	17	7.2 %
61–65	7	3.0 %
66–70	10	4.2 %
70+	6	2.5 %
Coach gender
Male	123	51.9 %
Female	113	47.7 %
Gender diverse	1	0.4 %
Coach ethnicity
European/Pākehā	179	75.5 %
Māori	38	16.0 %
Pasifika	7	3.0 %
Asian	9	3.8 %
MELLA	3	1.3 %
Missing	1	0.4 %
Sport
Badminton	14	5.9 %
Cricket	13	5.5 %
Football (soccer)	27	11.4 %
Hockey	16	6.8 %
Netball	49	20.7 %
Rugby union	27	11.4 %
Other sports	91	38.4 %
Payment type
Volunteer	173	73.0 %
Non-volunteer	59	24.9 %
Missing	5	2.1 %
Athlete age*
<5	7	3.0 %
5–7	26	11.0 %
8–10	64	27.0 %
11–13	108	45.6 %
14–16	138	58.2 %
17	75	31.6 %
Athlete gender
Female	93	39.2 %
Male	49	20.7 %
A mixture of both	95	40.1 %

*Participants can coach multiple age groups so the total percentage exceeds a hundred percent. MELAA: Middle Eastern, Latin American and African.

The survey was successfully delivered to 6037 email addresses registered with the Sport NZ “Balance is Better” newsletter. A total of 2206 people opened the email, and 517 people clicked to open the survey. In total, 237 people completed the survey. There were slightly more male (51.9%) than female participants, and one participant identified as gender diverse. Regarding age groups, the largest proportion of the participants were middle-aged, with more than half of the coaches in their 40s and 50s. A vast majority of the coaches were European/Pākehā² (75.5%), followed by Māori (16%). One in five participants were netball coaches, and about one in ten were coaching football and rugby, respectively. There were a similar number of badminton, cricket, and hockey coaches (5% − 6%). Most of the participants were coaching as volunteers (73%), whereas about a quarter were receiving some sort of “payment-in-kind” for their coaching efforts.

In terms of the characteristics of the participants’ athletes, most were aged between 11 to 16 years old. Three percent of participants had athletes younger than five years old, and 11% were coaching athletes between the age of five to seven. Forty percent of the participants stated that they coached a mix of male and female athletes. Another 40% had female athletes only. Twenty percent of the participants had male athletes only.

Measures and data collection

As this study was the first of its kind in the New Zealand context, an online Google survey was designed by the first and third author with the aim to examine New Zealand coaches’ understandings of CSG discourses, resources, and policy; and the types of pedagogical changes they perceived were necessary in response. Each question in the survey was formulated based on the overarching research question – of which stemmed from a review of international scholarship on the impact of CSG policy on coaching practice: (how) does CSG policy influence coaching behaviours in New Zealand? (See Appendix 1 for the full version of the questionnaire³). The draft survey was then sent to two CSG consultants for review and to improve face validity. In this process, several questions were identified as repetitive and removed, and suggestions were made to improve clarity and grammar. The survey was then reviewed by a Sport NZ CSG/policy consultant who provided feedback on the questions they felt were important for their work in disseminating CSG resources. Prior to dissemination, support was sought from a Faculty-based service (at the first author’s institution) that specialises in providing expertise in quantitative data analysis and research. The service provided advice on errors like double-barrelled, confusing, and leading questions.

Ethical approval was granted for this project by the lead author’s university Human Ethics Committee. Participants were required to read an information sheet and complete an opt-in consent form before commencing. The survey was distributed via a hyperlinked article featured in Sport NZ’s regular “Balance is Better” newsletter. The survey was first distributed on the 27^th May, 2021 and was closed on the 30^th August, 2021.

Data preparation and analysis

Responses to the Google survey were downloaded in an Excel spreadsheet and analysis was conducted using jamovi version 2.0. Data recoding was conducted to transform the demographic data from string to numeric data for the purpose of the subsequent data analysis. “Ethnicity” was recoded according to five ethnic groupings: Māori, Pasifika, European/Pākehā, Asian, and “Middle Eastern, Latin American, and African (MELAA)”. The “Athlete Age” variable allowed for multiple options. To recode this variable, six new dummy variables were created, one for each age group, with “1” indicating selected, and “0” indicating not selected. The variable “Sports” was recoded so that sports with less than ten participants were grouped together into “Other sport”. The variable “Payment Type” was also recoded into new variables, with the options of “No – not any payment at all” and “Yes, payment-in-kind (i.e. petrol vouchers, food vouchers, etc.)” recoded as “Volunteers” and “Non-volunteers” respectively.

In many cases, participants were invited to provide additional written answers to some questions. A content analysis of these qualitative responses was conducted by deductively coding data that demonstrated the influence CSG policy was having on participants (i.e. the research question), and examples of where they articulated being in danger, considered as dangerous, and/or were in a position to observe for danger (i.e. Bennett et al, 2022). Following Saldaña (2013) recommendations, the data were compiled in a spreadsheet, then read and re-read multiple times by the first and third authors for familiarisation. A process of iterative consensus validation, supported by frequent and in-depth communications between the researchers, ensured that the codes were placed in categories that made the most analytic sense (Keegan et al., 2009). This process involved deep and critical conversations that encouraged us to consider our positionality at each stage of the analysis. The analysis was completed when each member of the research team agreed that the presented responses were both salient to the research question and added important nuance to the quantitative findings. The qualitative responses are presented with the findings below to add additional depth and context to the quantitative data.

Results

Overall perceptions of child safeguarding policies

Coaches’ awareness of the policies and their perceptions of these policies were measured by five core questions in the survey. Table 2 provides an overview on the response patterns of all the participants across the five major questions. As can be seen from the table, 65.8% of the participants were aware of the existence of the policy, whereas 12.7% of the participants were unaware of CSG policies in their club/sport. There was an equal number of coaches who believed there was a need of more policy and those who did not believe so (about 27% respectively). There were more people who perceived the policy has helped their coaching practice (about 40%) compared to those who believed it was not helping (about 30%). The majority of the participants (about 80%) believed that the policy was not hindering their role as a coach, and 60% perceived that the policy has not changed what they did as a coach.

Table 2. An overview of the percentages of answers chosen for the five questions.

	Existence of policy	Need of more policy	Helping	Hindering	Changed
No	12.7 %	27.0 %	29.1 %	79.3 %	60.3 %
Maybe	21.1 %	38.8 %	22.8 %	6.8 %	9.7 %
Yes	65.8 %	27.4 %	39.2%	5.5 %	21.5 %
No comment	0.4 %	6.8 %	8.9 %	8.4 %	8.4 %

“No comment” was treated as missing data in the following data analysis. Little’s MCAR test was employed which showed that the data were missing at random (χ2 = 45.69, df = 33, p = 0.07).

Table 3 shows the descriptive statistics of the five core questions. The skewness and kurtosis data showed that none of the five variables are normally distributed, using the threshold of 3.29 for medium sized samples (Kim, 2013). Given that the five interested/dependent variables were measured at the ordinal level (i.e. No, Maybe, Yes), the data were treated as ordinal. The answer “No” was recoded as 0, “Maybe” was recoded as 1, and “Yes” was recoded as 2. Ordinal logistic regressions were employed to investigate whether the characteristics of the coaches and their athletes predicted the participants’ perceptions of and attitudes towards the CSG policies in their club/sport.

Table 3. Descriptive statistics of the five questions.

						Skewness		Kurtosis
	N	Missing	Mean	Median	SD	Skewness	SE	Kurtosis	SE
Existence of policy	236	1	1.53	2	0.71	−1.20	0.16	−0.01	0.32
Need of more policy	221	16	1.01	1	0.77	−0.01	0.16	−1.29	0.33
Helping	216	21	1.11	1	0.86	−0.22	0.17	−1.62	0.33
Hindering	217	20	0.19	0	0.53	2.68	0.17	5.97	0.33
Changed	217	20	0.58	0	0.85	0.94	0.17	−0.95	0.33

Are there child safeguarding policies in place in your sport/club?

As can be seen from Table 4, an increase in age (in year levels by 5) was associated with an increase in the odds of being aware of the existence of the policy – with an odds ratio of 1.201 (95% CI, 1.054 to 1.376) (Wald χ2 (1) = 7.589, p < .01). That is, older participants, when compared to younger participants, were more likely to be aware of the existence of CSG policy. The odds of badminton coaches being aware of the existence of CSG policy was about 92% to 94% higher than that of football, hockey, netball, and other sports coaches (odds ratio of 0.056 to 0.083). The odds of non-volunteer participants being aware of the existence of CSG policy was 2.965 (95% CI, 1.278 to 7.330) times that of volunteers (Wald χ2 (1) = 6.507, p = .01). In addition, the odds of participants who coached athletes aged 5–7 being aware of the existence of the policy was 3.777 (95% CI, 1.118 to 17.674) times that of participants who did not coach that age group (Wald χ2 (1) = 4.642, p = .05).

Table 4. Ordinal logistic regression model for existence of policy.

						95% Confidence Interval
Predictor	Estimate	SE	Z	p	Odds ratio	Lower	Upper
Coach age	0.183	0.068	2.701	< .01	1.201	1.054	1.376
Coach gender (Female)
Male	0.197	0.407	0.483	0.63	1.217	0.547	2.719
Coach ethnicity (European/Pākehā)
Māori	0.520	0.418	1.243	0.21	1.682	0.755	3.930
Pasifika	1.190	0.927	1.284	0.20	3.289	0.601	26.371
Asian	1.873	1.189	1.584	0.11	6.510	0.864	137.652
MELLA	−1.189	1.483	−0.802	0.42	0.304	0.011	8.403
Sport (Badminton)
Cricket	−1.663	1.417	−1.174	0.24	0.190	0.007	2.408
Football	−2.537	1.309	−1.938	0.05	0.079	0.003	0.730
Hockey	−2.837	1.313	−2.160	0.03	0.059	0.002	0.549
Netball	−2.885	1.292	−2.233	0.03	0.056	0.002	0.490
Rugby union	−1.044	1.352	−0.773	0.44	0.352	0.013	3.727
Other sports	−2.489	1.227	−2.029	0.04	0.083	0.004	0.616
Payment type (volunteers)
Non-volunteers	1.087	0.443	2.453	0.01	2.965	1.278	7.330
Athlete age lower than 5	−0.192	1.413	−0.136	0.89	0.825	0.057	22.400
Athlete age 5–7	1.329	0.684	1.944	0.05	3.777	1.118	17.674
Athlete age 8–10	0.139	0.426	0.326	0.75	1.149	0.504	2.705
Athlete age 11–13	0.468	0.339	1.378	0.17	1.596	0.829	3.157
Athlete age 14–16	0.184	0.387	0.475	0.64	1.202	0.561	2.572
Athlete age 17	0.336	0.395	0.850	0.40	1.399	0.651	3.087
Athlete gender (mixed)
Female only	0.243	0.400	0.608	0.54	1.275	0.582	2.812
Male only	−0.325	0.517	−0.628	0.53	0.723	0.259	1.989

Note. Categories in brackets are used as reference groups. Variables in bold are statistically significant at p = .05 level.

Is there a need for more child safeguarding policy in your sport/club?

As can be seen from Table 5, the odds of football, hockey, and other coaches perceiving more policy was needed was about six times that of badminton coaches (Wald χ2 (6) = 16.626, p = .01). In addition, the odds of participants who coached athletes younger than five years old perceiving more policy was needed was 8.226 (95% CI, 1.359 to 58.770) times that of other coaches who did not coach that age group (Wald χ2 (1) = 5.279, p = .03).

Table 5. Ordinal logistic regression model for need of more policy.

						95% Confidence Interval
Predictor	Estimate	SE	Z	p	Odds ratio	Lower	Upper
Coach age	0.005	0.056	0.088	0.93	1.005	0.900	1.120
Coach gender (Female)
Male	−0.332	0.328	−1.011	0.31	0.718	0.376	1.360
Sport (Badminton)
Cricket	−0.074	0.880	−0.084	0.93	0.929	0.161	5.280
Football	1.856	0.758	2.449	0.01	6.395	1.493	29.880
Hockey	1.907	0.773	2.469	0.01	6.736	1.531	32.500
Netball	1.314	0.723	1.816	0.07	3.720	0.929	16.310
Rugby union	1.179	0.767	1.536	0.13	3.250	0.741	15.410
Other sports	1.684	0.636	2.647	0.01	5.386	1.607	20.210
Payment type (volunteers)
Non-volunteers	−0.122	0.342	−0.355	0.72	0.885	0.452	1.740
Athlete age lower than 5	2.107	0.943	2.235	0.03	8.226	1.359	58.770
Athlete age 5–7	−0.880	0.468	−1.880	0.06	0.415	0.162	1.030
Athlete age 8–10	0.214	0.343	0.624	0.53	1.239	0.632	2.430
Athlete age 11–13	−0.004	0.278	−0.013	0.99	0.996	0.577	1.720
Athlete age 14–16	0.198	0.323	0.613	0.54	1.219	0.647	2.300
Athlete age 17	0.038	0.315	0.120	0.90	1.039	0.560	1.930
Athlete gender (mixed)
Female only	0.223	0.364	0.613	0.54	1.250	0.613	2.560
Male only	0.647	0.440	1.472	0.14	1.910	0.809	4.570

Categories in brackets are used as reference groups. Variables in bold are statistically significant at p = .05 level. Coach ethnicity was omitted from the model for the model to run successfully.

The prompt to “provide an example of where/why there is a need for more child safeguarding policy in your sport/club” garnered 108 qualitative responses – of which 22% noted that: clarity of procedures through regular discussion (i.e. between coaches/from administrators); training; and clearer guidelines were required. As an example, one participant commented that: “Not necessarily more [is needed] just turning what is already available into daily practice”. Other participants commented that: “You forget the rules”; and “I don’t have enough understanding of what safeguarding is actually about”.

Is the child safeguarding policy used in your sport/club helping you in your role as a coach?

As Table 6 shows, an increase in age (expressed in year levels by 5) was associated with an increase in the odds of perceiving the policy as helping with an odds ratio of 1.152 (95% CI, 1.024 to 1.300) (Wald χ2 (1) = 5.564, p = .02). In addition, the odds of non-volunteer participants perceiving the policy as helping was 2.177 (95% CI, 1.070 to 4.550) times that of volunteer coaches (Wald χ2 (1) = 4.618, p = .03).

Table 6. Ordinal logistic regression model for helping.

						95% Confidence Interval
Predictor	Estimate	SE	Z	p	Odds ratio	Lower	Upper
Coach age	0.142	0.061	2.334	0.02	1.152	1.024	1.300
Coach gender (Female)
Male	−0.005	0.341	−0.015	0.99	0.995	0.509	1.940
Sport (Badminton)
Cricket	−0.640	0.882	−0.726	0.47	0.527	0.090	2.930
Football	0.065	0.767	0.085	0.93	1.068	0.232	4.780
Hockey	−0.740	0.767	−0.965	0.34	0.477	0.103	2.120
Netball	−0.540	0.722	−0.748	0.46	0.583	0.137	2.370
Rugby union	0.605	0.778	0.777	0.44	1.831	0.392	8.450
Other sports	0.076	0.623	0.122	0.90	1.079	0.305	3.610
Payment type (volunteers)
Non-volunteers	0.778	0.368	2.116	0.03	2.177	1.070	4.550
Athlete age lower than 5	−0.181	0.973	−0.186	0.85	0.834	0.124	6.030
Athlete age 5–7	0.605	0.523	1.157	0.25	1.832	0.672	5.370
Athlete age 8–10	0.356	0.371	0.958	0.34	1.427	0.693	2.990
Athlete age 11–13	0.461	0.288	1.601	0.11	1.585	0.904	2.800
Athlete age 14–16	0.436	0.339	1.285	0.20	1.547	0.798	3.030
Athlete age 17	−0.176	0.332	−0.530	0.60	0.839	0.435	1.610
Athlete gender (mixed)
Female only	0.236	0.379	0.622	0.53	1.266	0.605	2.680
Male only	−0.647	0.457	−1.417	0.16	0.523	0.211	1.280

Categories in brackets are used as reference groups. Variables in bold are statistically significant at p = .05 level. Coach ethnicity was omitted from the model for the model to run successfully.

The prompt to “provide an example of how child safeguarding policy used in your sport/club helping you in your role as a coach” garnered 101 qualitative responses – of which 27% noted that the CSG policy that does exist in their club/sport provides “peace of mind” as participants acknowledged their safety was protected by the policy. For example: “[CSG policy] ensures my own safety and protection when alone coaching others’ children”. Similarly, 18% of responses suggested that the presence of CSG policy helps to establish boundaries (i.e. what is deemed appropriate behaviour) and expectations of coaches. For instance: “[CSG policy] gives me a base to work off and so both myself, my players and their parents are aware of the expectations”.

Is the child safeguarding policy used in your sport/club hindering you in your role as a coach?

The results showed that none of the coach demographics or the athletes characteristics were associated with perceptions on whether policy was hindering their role as a coach (Table 7).

Table 7. Ordinal logistic regression model for Hindering.

						95% Confidence Interval
Predictor	Estimate	SE	Z	p	Odds ratio	Lower	Upper
Coach age	−0.092	0.085	−1.088	0.28	0.912	0.770	1.080
Coach gender (Female)
Male	0.874	0.485	1.802	0.07	2.396	0.940	6.390
Payment type (volunteers)
Non-volunteers	−0.215	0.531	−0.405	0.69	0.807	0.267	2.180
Athlete age lower than 5	−0.831	1.314	−0.633	0.53	0.436	0.018	4.800
Athlete age 5–7	0.177	0.731	0.243	0.81	1.194	0.234	4.440
Athlete age 8–10	0.515	0.538	0.957	0.34	1.673	0.576	4.870
Athlete age 11–13	0.264	0.435	0.605	0.55	1.302	0.551	3.080
Athlete age 14–16	0.078	0.524	0.148	0.88	1.081	0.392	3.130
Athlete age 17	0.475	0.508	0.935	0.35	1.608	0.588	4.390
Athlete gender (mixed)
Female only	0.205	0.513	0.400	0.69	1.228	0.448	3.400
Male only	−1.066	0.720	−1.481	0.14	0.344	0.070	1.290

Categories in brackets are used as reference groups. Coach ethnicity and sport were omitted from the model for the model to run successfully.

The prompts to “provide an example of how child safeguarding policy used in your sport/club hinders you in your role as a coach” garnered 46 qualitative responses – of which 15% noted a degree of worry or concern about the CSG protocol in their club/sport. This was phrased in the following ways: “It’s sometimes hard to know the line around being appropriately caring/empathetic and showing no emotion or putting yourself in innocent but potentially compromising situations – e.g. a child hurts themselves and wants a hug”. Additionally, one participant commented that: “At times you are more conscious of following protocol and [avoiding] risk, than actually coaching”.

This sentiment was further expressed with specific mention about gender of the coach:

… there can be a problem getting men on board for the threat of being vulnerable or put in tricky situations with female competitors. Can be difficult for women being involved in male codes too.

Being a Male Coach/Instructor, teaching males is fine, but as soon as you are teaching females you have to take a step back and look at how you coach in the eyes of parents and random passers by to make sure nothing looks weird even though in the sport you are coaching everything is normal.

Has the child safeguarding policy used in your sport/club changed what you do as a coach?

Coach age was the only significant predictor of participants’ perceptions on whether CSG policy had changed what they do as a coach (Table 8). An increase in age (expressed in year levels by 5) was associated with an increase in the odds of perceiving the policy has changed what coaches do with an odds ratio of 1.118 (95% CI, 1.004 to 1.250) (Wald χ2 (1) = 4.137, p = .05).

Table 8. Ordinal logistic regression model for changed.

						95% Confidence Interval
Predictor	Estimate	SE	Z	p	Odds ratio	Lower	Upper
Coach age	0.112	0.056	2.007	0.05	1.118	1.004	1.250
Coach gender (Female)
Male	−0.219	0.336	−0.652	0.52	0.803	0.414	1.550
Coach ethnicity (European/Pākehā)
Māori	0.688	0.385	1.786	0.07	1.990	0.928	4.230
Pasifika	1.394	0.781	1.784	0.07	4.030	0.830	18.830
Asian	1.104	0.630	1.752	0.08	3.015	0.844	10.300
MELAA	−0.153	1.236	−0.124	0.90	0.858	0.039	8.170
Sport (Badminton)
Cricket	−0.082	0.826	−0.100	0.92	0.921	0.177	4.680
Football	0.634	0.755	0.840	0.40	1.886	0.437	8.580
Hockey	−0.446	0.826	−0.540	0.59	0.640	0.119	3.200
Netball	−0.198	0.748	−0.265	0.79	0.820	0.192	3.680
Rugby union	0.143	0.774	0.185	0.85	1.154	0.255	5.410
Other sports	0.283	0.605	0.468	0.64	1.327	0.418	4.600
Athlete age lower than 5	0.690	0.890	0.776	0.44	1.995	0.334	11.430
Athlete age 5–7	0.348	0.490	0.710	0.48	1.417	0.521	3.640
Athlete age 8–10	−0.610	0.367	−1.663	0.10	0.543	0.260	1.100
Athlete age 11–13	−0.080	0.286	−0.279	0.78	0.923	0.526	1.620
Athlete age 14–16	0.019	0.315	0.060	0.95	1.019	0.549	1.900
Athlete age 17	0.376	0.307	1.224	0.22	1.457	0.796	2.660
Athlete gender (mixed)
Female only	−0.164	0.370	−0.444	0.66	0.849	0.408	1.750
Male only	0.016	0.455	0.036	0.97	1.016	0.416	2.490

Categories in brackets are used as reference groups. Variables in bold are statistically significant at p = .05 level. Payment type was omitted from the model.

The prompt to “provide an example of how child safeguarding policy has changed your practice as a coach in your sport/club” garnered 74 qualitative responses – of which 22% of participants indicated that they had changed their coaching pedagogy in some form in an effort to take care to avoid risk. For instance: “I’m much more aware of how actions might be misinterpreted or considered inappropriate, e.g. asking, when introduced to a new participant, if they are comfortable shaking hands”. Additionally, one participant commented that: “Previously we didn’t ask for permission to touch someone when correcting their form, now we do”.

In contrast, 20% of the qualitative responses indicated that changes in coaching pedagogy were influenced by a heightened understanding of athlete wellbeing in the context of sport training. For instance: “I am more aware of balance, injury prevention looking for signs of distress”. Also: “In the old days I might be a little tougher on young people than I am today but, [nowadays] I’d make sure to debrief with [the athlete] after and perhaps chat with their parents about it”.

An additional 20% of the qualitative responses indicated that participants had changed their pedagogy in ways that they felt less connected to athletes in comparison to past experiences: “I’ve had to become less connected with both genders that I coach. Less connected with the females so it doesn’t look weird to outsiders, but then less connected with the males so that there isn’t any imbalance or favouritism showing”. Responses indicated that participants’ limited physical interactions (i.e. the avoidance of touch), and/or one-on-one instruction were examples of changes to pedagogy.

Discussion

Results indicated approximately two thirds (65.8%) of participants were aware of CSG policy in their club/sport. Conversely, one in five (21.1%) participants indicated that they were unsure about the existence of CSG; and slightly over one in 10 (12.7%) participants signalled that they were unaware of any relevant CSG policies in their club/sport. In essence, findings indicated that approximately one third (34%) of participants are either unaware or uncertain of what CSG policies are in place to guide their coaching practice. More participants perceived that the CSG policy in their sport/code has helped their coaching practice (about 40%) compared to those who believed it was not helping (about 30%). Many of the participant coaches (about 80%) believed that the policy was not hindering their role as a coach. However, approximately two in three participants (60%) perceived that the policy has not changed what they did as a coach.

In light of these findings, we theorise that the apparent contradiction between participants’ awareness of CSG policy, and the indication that they have not changed their practice in response, may be closely related to an overall level of uncertainty. That is, coaches are aware that some type of response to CSG policy is required, however, they are uncertain about how to go about making changes in their practice in ways that are consistent with guidelines and the realities of their sport. As an example, it is possible that efforts by Badminton NZ to introduce comprehensive CSG training for coaches (Sport, 2022b) resulted in the increased odds that badminton coaches were aware of CSG policy in comparison to other sports (thus less likely to see a need for more guidance), and as a result, had a clearer idea of how to apply CSG policy to their coaching. Indeed, if this was the case, efforts to mitigate discrepancies between “best practice” and “best guesses” in other areas of the sport sector should include ongoing and nuanced coach education.

Notwithstanding the efforts made thus far to advocate for, and evolve “best practice” in CSG in New Zealand vis-à-vis guidance from safeguarding consultants, forums, webinars, and online CSG modules (Sport NZ, 2022a; J. Richards, personal communication, 19 April 2022), the findings of this study highlight a number of noteworthy concerns that policy writers and coach developers could. Below, we further discuss the findings to examine how characteristics such as age, gender, ethnicity, sport, and volunteer status influenced perceptions of CSG policy and coaching pedagogies.

Coach age

Statistical analysis indicated that, as participant age increased, so too did awareness of the existence of CSG policy. Further, as participant ages increased, there also appeared to be a tendency for these cohorts to change their coaching behaviour as a result of CSG policy; and to view CSG policy as helpful in their coaching role.

In our search for a possible interpretation of these findings, we consider Schildberg-Hörisch’s (2018) proposition that older participants are more risk averse than their younger counterparts (ergo are more inclined to be aware of CSG policy). Awareness and adoption of CSG policy in education/coaching practice has previously been positioned in terms of danger, and educators’/coaches’ increasing recognition of danger or risk when working with young people. This includes, as Bennett et al., (2022) concluded; 1) the coach being viewed as dangerous; 2) the coach in danger of accusation; and 3) the coach as an observer of danger (see also Beck et al., 1992; Giddens, 1999; Lang & Pinder, 2017; McWilliam & Jones, 2005; Piper & Smith, 2003). As the older cohort may be, according to Schildberg-Hörisch’s (2018) proposition, more inclined to be aware of the potential areas of dangers inherent in coaching youth athletes, this notion may go some way in explaining the increased recognition of CSG policy (and its perceived helpfulness). Nevertheless, an alternative interpretation could be that younger cohorts are more familiar with CSG, for instance, as a result of their own experiences and/or CSG dissemination and education efforts. Either way, this data presents thought-provoking challenges for sport organisations and writers of CSG policy when attempting to engage younger coaches, who, according to our interpretations of the data, are less risk averse and less aware of CSG policy.

The relationship between awareness of CSG policy, age and experience of coaches, risk aversion levels, and the impact that one or more of these factors may have on pedagogical behaviour requires further exploration. Indeed, to account for the array of possible explanations in this line of enquiry, a qualitative approach with this older cohort of participant coaches may allow for important nuances to emerge for researchers, coach developers, and policy writers to consider.

Athlete age

Analysis of athlete ages yielded noteworthy findings regarding awareness of CSG policy. For instance, the odds of participants who coached athletes aged between 5–7 being aware of the existence of CSG policy was almost four times that of participants who coached older age groups. While encouraging (and perhaps unsurprising given the obvious vulnerability of children aged 5–7), levels of awareness appear to reduce as athlete age increases. We see this as an important finding, particularly as New Zealand’s Children’s Act (Ministry of Education, 2015) defines “children” as those up to the age of 18 years.

We posit that participants’ awareness levels may be influenced by time pressures and organisational limitations (i.e. lack of human or financial resources needed to distribute and promote policy) experienced by the primarily volunteer workforce (see below). Therefore, Sport NZ, National Governing Bodies (NGBs), and in turn, National Sport Organisations (NSOs) may need to (re)consider how they support coaches to engage with/become aware of current CSG policy across athlete age levels. Indeed, this may require more targeted/athlete-age-specific resources (particularly as 57% of participants coaching athletes under 5 years old were still more likely to indicate a need for more policy). Moreover, dissemination efforts must also remain mindful that “[forgetting] the rules” and “a lack of understanding of what safeguarding is actually about” appear to be barriers for participants to implement CSG procedures in their coaching practice.

Volunteers

The odds of non-volunteer coaches being aware of the existence of the policy was nearly three times that of volunteer coaches. Additionally, 55% of paid coaches felt that current CSG policies were helpful to them in the coaching role, compared with only 33% of volunteer coaches feeling the same way – albeit, neither cohort were more or less likely to change their practice in response to CSG policy.

Internationally, Piper et al. (2012) have highlighted the issues surrounding volunteer coaches’ uptake and implementation of CSG. Specifically, the lack of dedicated training in the area of child welfare results in an inconsistent application of CSG policy and procedures. Approaching duties with the same vigour of their paid contemporaries, volunteers bring with them a number of preconceptions and/or lack opportunities to engage in additional training on the matter of safe conduct (Piper et al., 2012 Compounding this is the numerous tasks required of volunteers in community sport clubs (e.g. committee roles, fund raising, facility management, and coaching, etc.). This is an important distinction from other contexts where the non-volunteer – for example, a teacher or professional coach – (potentially) has a number of working and training hours dedicated to the processes and procedures required to manage risk.

Further insights from the quantitative analysis revealed that while paid coaches were both more aware and more inclined to see existing CSG policy as helpful, neither group were more or less likely to change their practice in response to CSG policy. However, when considering the qualitative responses to these data we see contradictions. For example, qualitative data suggested that participants were making a “conscious” decision to avoid risk: “[a]t times you are more conscious of following protocol and avoiding risk, than actually coaching”; with others stating that “[i]t’s sometimes hard to know the line around being appropriately caring/empathetic and showing no emotion or putting yourself in innocent but potentially compromising situations – e.g. a child hurts themselves and wants a hug”. Thus, despite the quantitative data indicating a limited influence on participants’ coaching practice overall, the qualitative responses indicated that those who had made changes, have done so based on assumptions or (mis)interpretations of CSG policy.

Here, we are drawn to Denison’s (2019) warning that coaches are constantly exposed to contexts of social construction. That is, coaches are sculpted by dominant social narratives that act to subconsciously shape the way they think, speak, act, and therefore, coach. This strong socialising process weaves itself deeply into the fabric of society and normalises itself across a wide range of socialising institutions, including sport and education (Coakley, 2021). As Fletcher (2013) and Johnson (2000, 2013) explain, current CSG policy appears to be driven by a “moral panic” where coaches are seen as potentially dangerous and children exposed to coaches are all potentially at risk. It appears that participants in our study may think and possibly act differently as a result of CSG policy – a change that is potentially driven by dominant media narratives that privilege and highlight bad behaviour of coaches (e.g. George, 2020; Hurley, 2019; Pesta, 2019).

Regardless of the classification of coaches as voluntary or non-voluntary there appears to be a limited amount of research exploring the impact that CSG policy, or the changing social narrative around child safety in sport, is having on the way people coach. Given our interpretations, we advocate that future CSG policy development and dissemination efforts carefully consider the increasing amounts of research literature in this area – and specifically, on the impact that misinterpretations or privileged social narratives may have on the coaching profession. We continue this discussion in the “Further Discussion” section below.

Gender and (mis)interpretations

Findings indicated that 22% of the study participants coaching male athletes (compared to 14% of people coaching female athletes) are more inclined to have changed their practice as a result of current CSG policy in their sport/club. Importantly, however, when viewed according to coach gender, results indicated that 24% of male coaches (compared to 20% of female coaches) were more inclined to change their coaching practice in response to CSG policy. Our quantitative data analysis revealed that the nature of these changes related to restricting and/or cautioning against one-on-one interactions between coaches and athletes. For instance, participant comments alluded to approaches that emphasised: “[t]wo (adult) coaches in gym at all times. [You should] never be alone with an athlete” and “[you must ensure] children are accompanied by a parent or caregiver, [so that the] coach is never alone with a child”.

The guidelines for child protection policies in New Zealand encourages “open door” policies (i.e. restricting one-on-one time with athletes), and encourages organisations to consider the inclusion of statements that outline expectations for staff regarding the need for one-on-one coaching (Ministry of Education, 2016). In this sense, while these may appear to be reasonable responses from participants, it is the nature of the language used by participants that indicates a degree of (mis)interpretation (eg. “never be alone … ”) that is not evident in guiding documents (Ministry of Education, 2016).

Moreover, our analysis indicated a tendency for interpretations to be made according to the gender of the coach and/or athlete. For instance, some participants suggested that the CSG policy they were aware of related to the restriction of male coaches’ one-on-one interactions with female athletes. For example, “Male staff are involved as long as appropriate then exit when it is ‘women’s business”., “Males aren’t allowed to be alone in confined areas with female athletes”, and: “For NZ team trips there are requirements for … gender matching to athletes (i.e. if there is a male athlete in the team there must be a male on team management staff)”.

Importantly, gendered separation does not appear in the literature as “best practice”, nor does it appear to be advocated by Sport NZ’s CSG resources (Sport NZ, 2022a). Similar in nature to coaches’ well-intentioned (mis)interpretation of CSG policy witnessed in the United Kingdom (Lang & Pinder, 2017; Piper & Smith, 2003; Piper & Stronach, 2008; Piper et al., 2012), scholars have previously troubled the ways in which this can result in problematic “cultures of suspicion” to the detriment of the coach-athlete experience (Lang & Pinder, 2017). Certainly, participants’ comments about “[b]eing a Male Coach/Instructor … [requires a] step back and look at how you coach in the eyes of parents and random passers-by to make sure nothing looks weird even though in the sport you are coaching everything is normal” offers an example of how this problematic trend is unfolding in New Zealand’s sport sector. As one participant commented, such trends may also lead to: “problem[s] getting men on board for the threat of being vulnerable or put in tricky situations with female competitors”.

As scholars have previously suggested, increased frequency of discourse around the issue of child abuse has caused educators to reorganise elements of their practice (Lang, 2010, 2015; Öhman & Quennerstedt, 2017; Öhman, 2017). Tait (2001) explains that the first rationale for this, needless to say, is for the protection of the children engaged in physical activity (by forbidding educators to be alone with students, introducing strict protocols surrounding incident reporting, a range of awareness programmes, and “no-touch” policies). Tait (2001) suggests that the protection of the educator and the institution (i.e. the school, club, or team) he/she operates within constitutes a second rationale, forcing educators “into action to protect themselves” (p.2) – specifically, through actions or interpretations of guidelines that limit or restrict physical contact between educator and learner. Ironically, the consequences of these interpretation(s) “ … may in turn produce an array of other ‘risk’ groupings [such as] ‘male teachers at risk of leaving the profession’, ‘children at risk of fearing all physical contact’, ‘boys at risk of having no gentle male role models’ and ‘schools at-risk of witch-hunts’ … ” (Tait, 2001, p. 13; see also Garratt et al., 2013).

The data presented above provides an indication that – although approximately two thirds of participants are aware of, and have changed their coaching practice in light of CSG policy – the potential adoption of defensive pedagogies as a result of unproblematic misinterpretations of CSG policy that requires further attention. In the final section of this article, we unpack this notion of defensive pedagogies further before offering our concluding remarks.

Further discussion – defensive pedagogy

Öhman and Quennerstedt (2017) suggest that the discourse surrounding CSG often places the spotlight on a narrow interpretation of the child’s rights. Namely, the child’s right to protection. However, primary focus on the right of the child to protection, in an effort to minimise the risk to the child, can result in policies and procedures that are geared towards the avoidance of risk. In the case of the current data, for instance, this can be seen in the separation of male coaches from female athletes as a defensive pedagogical strategy.

Unsurprisingly, the meta narrative of child protection discourses define the vocabulary and context in which CSG issues are understood, and how policies are shaped (Fraidin, 2010). As Nelson and Knudsen (1986) have suggested, child abuse elicits strong, fairly uniform emotional response – that is, there is little public support for abusers and huge support for developments in CSG policy. As Lang and Pinder (2017) suggest, however, this serves to represent only one (dominant) version of many possible accounts of an interaction between a coach and an athlete. We, like Öhman and Quennerstedt (2017), are mindful not to understate the importance of efforts to minimise abuse within sport settings. However, if the consequences of this “meta” narrative (i.e. focusing on risk and danger) are not considered in depth, other possible interpretations and outcomes of the issue are ignored. In our view, and in line with Tait (2001), this “at risk” focus has paradoxically highlighted other “at risk” groupings in New Zealand’s sport settings – such as framing male educators working with female athletes; volunteers; and youth coaches as potentially risky or dangerous. Again, as Tait (2001) suggests, the irony is that the risks of male educators leaving the profession, and children learning to view inter-generational physical contact in educational settings as unusual are also concerning.

In particular, we are troubled by findings that indicate current interpretations and operationalisations of CSG policy are resulting in the unforeseen adoption and normalisation of certain defensive pedagogies. This trend can lead to the denial of children’s rights to natural, holistic and humanly interactive sport systems (Varea & Öhman, 2022; Öhman & Quennerstedt, 2017). As we suggest above, such (mis)interpretations are conceivably a result of a moral panic (Johnson, 2000) that exaggerates, and over time normalise defensive pedagogy as the dominant narrative in coaching circles – that is, in absence of robust critique or debate about the impact this has on the coach and athlete experience. Particularly in youth sport, and within a predominantly volunteer coach workforce. In this sense, we are mindful of McWilliam and Sachs (2004) suggestion that wholesale adjustments in the way(s) people coach – such as limiting male and female interactions in sport – are a result of (mis)interpretation and uncertainty that may not be “pedagogically appropriate or desirable” (p. 21). Especially when various sport and/or human movement contexts call for nuanced practices and pedagogies (Öhman & Quennerstedt, 2017).

Concluding thoughts

Despite the multifaceted efforts to disseminate and promote CSG cultures into the wider sport sector in New Zealand (i.e. online and face to face learning opportunities or experiences), the results of this study suggest that there is still work to be done. We suggest that policy and curriculum writers in New Zealand’s sport sector must consider the impact and uptake of dominant social narratives on CSG policy and practice. Specifically, this relates to the unproblematic adoption and conceptualisation of CSG policy that privileges the narrative dominated by a “moral panic”, and an uncritical uptake of “risk” discourses. As we have argued, it is possible that these underlying (mis)conceptions may be changing the way people coach in a manner that is not beneficial to the intergenerational relationship between coach and athlete.

Our final thoughts consider that coach development learning contexts and experiences should provide the opportunity to expose more coaches to notions of CSG – with guidance from CSG experts – that better ensure nuanced application and understandings. That is, “critical” coach development (Fyall et al., n.d.), where learning becomes a collaborative, active, authenticated experience that allows coaches to consider the role that power plays in the production or reproduction of knowledge and dominant social narratives. This enables coach developers, and those whom they are developing, the ability to interrogate the way they think, speak and coach. In essence, these developmental spaces allow for the interrogation of dominant social narratives that pervasively influence, and are unwittingly embedded in the ways we act. Potentially, this will allow coaches to discuss the application of CSG procedures in sport-specific settings with experts; whilst also considering pedagogies that are aligned with athlete age, gender, and competitive level. Importantly, it is critical for the actors in these spaces to consider whether (mis)interpretations of CSG are leading to the uptake of defensive pedagogies that may be unhelpful and detrimental to the development of children. Obviously, this presents a future line and direction of research, but one that will hopefully support a more conscious and positive conceptualisation and operationalisation of CSG policy.

Limitations

The results reported in this study should be considered in light of some limitations. First, as our search revealed no existing survey suitable to answer our research question, we designed the tool used in this study ourselves. Although this was done in close collaboration with CSG and quantitative research experts, our survey has yet to undergo wider use and testing. While this presents a limitation, it also provides an opportunity for future research – for instance, in other countries in the early stages of CSG policy development and dissemination – to further test and validate the survey. Similarly, future research should seek to obtain larger sample sizes towards more generalisable findings. Although our sample size of 237 coaches provides an initial snapshot, increased sample sizes would provide policy writers and coach developers with a clearer picture of the current understandings and uses of CSG policies in the New Zealand sport sector. Concurrently, increased sample sizes and focus on specific sports could highlight gaps in the resources and training capacity of different codes and NSOs. Such findings would provide nuanced insights that could lead to targeted and/or sport-specific CSG guidance in the areas of the sector that are struggling to keep up with best practice.

Acknowledgments

The authors wish to acknowledge Justin Richards (Associate Professor in Physical Activity and Wellbeing, School of Health, Victoria University of Wellington; Senior Evaluation Advisor, Sport NZ) and Josh Margetts (Senior Project Consultant - Sport Integrity, Sport NZ) for their assistance with and feedback on this project.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the University of Auckland

Notes

1. A teacher or other adult responsible for children in place of a parent or legal guardian.

2. A New Zealander of European descent as opposed to a Māori person.

3. The questionnaire was not a scale, and as such, there are no latent factors/structures.

Appendix

Are there child safeguarding policies in place in your sport/club?

Yes

No

Maybe

No comment

Please provide an example of child safeguarding policies or practices used in your sport/club.

Long answer text

On average, how often are the child safeguarding policies used in your sport/club referred to? (Click all that apply)

Daily

Weekly

Monthly

During induction

When a policy has been breached

Never

Don’t know

Other…

Is there a need for more child safeguarding policy in your sport/club?

Yes

No

Maybe

No comment

If applicable, please provide an example of where/why there is a need for more child safeguarding policy in your sport/club.

Long answer text

Is the child safeguarding policy used in your sport/club helping you in your role as a coach?

Yes

No

Maybe

No comment

If applicable, please provide an example of how child safeguarding policy used in your sport/club helping you in your role as a coach.

Long answer text

Is the child safeguarding policy used in your sport/club hindering you in your role as a coach?

Yes

No

Maybe

No comment

 

If applicable, please provide an example of how child safeguarding policy used in your sport/club hinders you in your role as a coach.

Long answer text

Has the child safeguarding policy used in your sport/club changed what you do as a coach?

Yes

No

Maybe

No comment

If applicable, please share an example of how child safeguarding policy has changed your practice as a coach in your sport/club.

Long answer text

Who is protected by the child safeguarding policies/procedures used in your sport/club? (Click all that apply)

The child(ren)/athlete(s)

The coach(es)

The club/sport

Managers

Parents

Don’t know

No comment

Other…

How do you access the child safeguarding resources used in your sport/club? (Click all that apply)

Sport NZ resources (i.e. Safe Sport for Children, online portal, etc.)

National Sporting Organisation (NSO) resources/workshops

Regional Sports Trust (RST) resources/workshops

’In house’/club/sport specific resources/workshops

NZ Police resources

External providers resources/workshops

Don’t know

Other…

What else, if anything, would you like to be covered in child safeguarding policy/measures used in your sport/club?

Long answer text

Finally, do you have any other messages you would like to share with policy makers or the wider public about how you feel on the matter of child safeguarding in sport?

Long answer text

Would you be willing to participate in an interview to offer further comments about how child safeguarding has/has not influenced your coaching?

Yes

No

If yes, please provide a contact email address/phone number

Short answer text