Community sports as a conduit for police–youth relationships: a scoping review

A key mission of the police is to instill public confidence in their role and to enhance their institutional legitimacy, fostering trust and cooperation with the community (Skogan and Hartnett, 1999). To this end, community policing, characterized by its emphasis on relationship-building and collaboration with local citizens, has proven to be an effective approach in increasing police legitimacy and shaping public perception of police in a positive way (Crowl, 2017; Mazerolle et al., 2013). However, the impact of community policing is not uniform across all areas and demographic groups, such as at-risk youths. Although there is evidence that community policing projects positively affect youths’ perception of police positively (e.g. Hinds, 2009), at-risk youth often exhibit lower levels of trust in law enforcement (Gau and Brunson, 2010) and are, because of this lacking trust, particularly challenging to engage (Kammersgaarden and Kyvsgaard, 2019). Despite the broader success of community policing, its effectiveness in reaching at-risk youths remains limited, highlighting the need for innovative and creative ways for police to connect with this target group. In this scoping review, we are interested in one platform that holds great potential for such ends: sports.

Sports have been incorporated extensively into social policy to combat a variety of social (youth) issues (Lawson, 2005), and one of the most touted claims of sports is that it represents an arena for unique meetings between diverse individuals (Bailey*, 2005), such as that of police officers and at-risk youths. The study of sport for such purposes has grown into an academic subfield, commonly referred to as sport-for-development (SFD). SFD has been extensively researched in the context of social integration and social capital, often being linked to civic participation and both interpersonal and institutional trust (Brown et al., 2014; Burrmann et al., 2020; Di Bartolomeo and Papa, 2019), and SFD’s positive benefits for at-risk youths have also been the subject of intense academic investigation (Newman et al., 2021; Spruit et al., 2018). However, it is abundantly clear that few studies focus on how sport can function as an innovative arena for police to engage hard-to-reach youths and build meaningful relationships, and it is obvious that little contact has been made between SFD and (community) policing scholarship. By researching this intersection, we hope to contribute to the different ways in which community policing can generate healthy relationships between police and youths. There is, to the authors’ knowledge, no review that focuses on the intersection between youth and police in the sport context whatsoever.

In summary, in this paper, we conduct a scoping review of the empirical literature on police–youth relationship building within the sport context. The overarching purpose of this review is to understand how positive contact between police and youths is facilitated through sports, what effects these interventions show and examine the moderating factors that impact the police–youth relationship within this context. The review’s research questions are as follows:

By answering these research questions, we seek to provide the first steps in integrating the principles of SFD with community policing strategies. The paper proceeds as follows. Firstly, we discuss community policing and the central tenants within community policing, including that of youth programs. Secondly, we offer a primer on SFD and then give a brief historical overview of sports within community policing. We then present the methodology, after which the results are presented. We finish with a discussion on how the results contribute to a broader understanding of community policing and its intersection with sport.

The concept of community policing emerged in the 1960 and 1970s as a response to the perceived deterioration of relations between the police and society (Skogan and Hartnett, 1999). Conceptually, community policing shifted the focus away from traditional and reactive policing to a more proactive approach that centers on relationship-building and partnership with the local community in order to address the root causes of social issues and crime (Trojanowicz and Bucqueroux, 1990). Theoretically, this reform movement was supported by new approaches in police science that focused on innovation and change, emphasizing the “coproduction” of public safety (Weisburd and Braga, 2019). Community policing as a concept has been difficult to define but can be broadly viewed as a four-dimensional concept, including a philosophical, a strategic, a tactical and an organizational dimension (Cordner, 1997). Unifying concepts running through these dimensions include that police officers should be more anchored in their local contexts, provide a broader support function than traditional policing does, be more attentive to the local community need and engage in a type of co-production of safety with their local surroundings (Skogan, 2006).

However, a persistent issue that community policing has faced is that, in order to collaborate with the local community, there needs to be a starting point in terms of trust. Unfortunately, many areas targeted for community policing are characterized by their notoriously low belief in authorities (Pino, 2001). A significant critique toward community policing has thus been that it does not operate with desirable effects in the places (and with the population) that needs it the most (Van Craen, 2013). For instance, in a large-scale experiment in six crime-ridden areas in the Global South, Blair et al. (2021) found that community policing had no effect on either trust or crime reduction. Similarly, surveying 1,300 US cities, Rukus et al. (2017) found that community policing worked well in low-crime communities but was not effective in high-crime communities.

Within community policing, much focus has been placed on police-youth interaction and how youth programs can assist in fostering positive interactions between the two (Anderson et al., 2007), especially those who are at risk of delinquency or disengagement from societal institutions. Youth programs, occasionally involving sports, are the conceptually closest we have in which sports and leisure are integrated into research on police–youth interaction and police legitimacy. Typical features of youth programs include being community-based, focusing on developing relevant competencies and important relationships and relying heavily on mentoring aspects (Roth et al., 1998). Considering the latter, youth programs have often connected with local police departments to provide a venue for youth and police to interact and to mitigate possible poor perceptions of one another. Notable initiatives include, for instance, Boys and Girls Clubs, Police Athletic Leagues and Police Explorer Scouts.

Generally, results from youth programs are positive. Thurman et al. (1993) found improved perceptions among both youths, parents and police after an eight-week program. Miner-Romanoff (2023) found that such programs facilitate a better view of police among youths in challenging neighborhoods, and Anderson et al. (2007) found that youths who were most characterized by social vulnerability gained the best effects from youth programs. Considering the general critique toward the effectiveness of community policing in underserved areas, Miner-Romanoff (2023) and Anderson et al.’s (2007) findings are particularly important. The perhaps most known initiative, to a broader audience outside of police science, is that of Midnight Basketball. Midnight Basketball gained great attention first in 1989 (Hartmann, 2001) and arranged police-supervised games across the US, targeting at-risk youth with the rationale that utilizing sports in these areas and during these late times would factor in geographical and temporal aspects that correlated with increased crime rates and serve as a diversion (Hartmann and Depro, 2006). Analyses by Hartmann and Depro (2006) showed that these initiatives did seem to have a crime preventative effect, but no analysis on police legitimacy or trust was ever done.

Notably, although a range of youth programs have seemingly generated positive effects, there have also been struggles to uphold these initiatives. Some examples include youth programs in Baltimore City that have experienced a discontinuity, partially based on political will and perceptions of what police officers “should” do (Caldas et al., 2018). Finally, despite tentative evidence of youth programs’ effectiveness, there is still a need to evaluate police-and-youth programs (Fine et al., 2021). According to Goodrich et al. (2014, p. 56), although attitudes between police and youth have been in focus, “… little research has focused on […] programs designed to improve the attitudes of police and youth toward one another.” We view this scoping review as one way to further identify and evaluate the effects of initiatives asking youth programs.

SFD has, since its inception around 2000, grown as a burgeoning discipline. One of the first steps in solidifying this field was the creation of the United Nations Office for Sport for Development and Peace, which provided a “… significant step towards official recognition and legitimacy for SFD” (Schulenkorf, 2017). Since then, SFD has been consistently backed by major authorities and actors such as the UN, the International Olympic Committee, the Commonwealth Games and many more (Kidd, 2008), to the point where sport participation, due to its assumed positive benefits, has been institutionalized as a human right in several important charters, including the UN’s Universal Declaration of Human Rights, the Council of Europe’s Sport for All Charter and UNESCO’s International Charter of Physical Education (Veal, 2023). Consequently, the field has grown exponentially stronger, and one explanation for this is sports’ cost-effectiveness and ability to “hook” a large number of people, including youths, for wider policy purposes (Schulenkorf and Sherry, 2021).

SFD has been generous in its approach, focusing on many types of social issues, such as immigration, disability and community development, and this literature has placed particular emphasis on engaging underserved, at-risk and marginalized youth, especially in relation to crime prevention and social reintegration efforts (Blomqvist, 2022, 2024). A significant body of SFD research explores how sport-based programs can serve as effective tools for diverting young people away from criminal activities and reducing antisocial behavior (Cameron and MacDougall, 2000; Morgan et al., 2020). Notably, the assumption about sports’ benefits is firmly grounded in a belief that sports educate youths in a myriad of ways. Some of these assumptions include moral development and a thorough understanding and respect for rules, boundaries and a “know-how” in relation to important institutions and authorities such as the school (Coakley, 2011) or the police, in this case. The two most frequently employed perspectives here are those of positive-youth-development and social capital. The former positioning the youth as a capable individual with growing potential, where the sports milieu works as a catalyst for this transformation (Holt et al., 2016). In this sense, crime prevention occurs through youths’ enhanced psychosocial competencies. Social capital emphasizes the importance of establishing healthy social relationships through sport and thereby acquiring important role models, peers and adults that constitute a healthy community (Schulenkorf and Sherry, 2021). Regardless of theoretical entry, to facilitate this transformation, much thought has been given to program design and delivery. A consistent key finding is that the responsible adults within the sporting context are of immense importance (Coalter, 2013). At the structural level, it is also noticeable that a significant benefit of using sports for social intervention is that sports movements constitute some of the largest popular movements existing (e.g. Fahlén and Stenling, 2019), making sport an easily accessible and cheap intervention for youth. Much like community policing, SFD actors also actively advocate for cross-sector collaboration to address complex social issues. Recognizing that the reach of sport alone has limits, SFD programs often partner with schools, social services and community organizations to create more comprehensive support networks for youth (Whitley et al., 2022).

To conclude this section, some principal affinities and differences between SFD and community police research warrant attention to understand how they may complement each other. At the outset, both literature use relational approaches and advocate for cross-sectoral cooperation but operate in different settings and with different authority structures. Additionally, and the most important difference, is that SFD research generally focuses on positive youth development and developing youths’ competencies but is void of research on how youths, through sport, develop attitudes toward the police. In contrast, the aspect of attitudinal improvement is a key theme in police science. However, secondly and interrelatedly, whereas this may be a key theme in police science, police–youth programs only shallowly analyze the sporting context in itself. SFD research is heavily based on the notion that the sporting milieu holds both promises and perils and needs significant attention for developmental outcomes to occur. Bringing the two together can hopefully contribute to a new focus in SFD (attitudes toward police or other authorities) and a more thorough examination of sport as an arena where attitudinal improvement can occur for police science.

In this article, we have adopted the scoping review approach (Arksey and O’Malley, 2005). The scoping review approach was deemed feasible because of the void of research on sports associations’ interactions with police and law enforcement agencies, in which we first sought to descriptively analyze the state-of-the art on the subject matter. We followed Arksey and O’Malley’s (2005) five steps in conducting our scoping review, with the addition that we also followed the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols Extension for Scoping Review.

Prior to conducting the full search, extensive preliminary searches were undertaken by the team to carve out keywords and understand the landscape, and in particular, the intersection between policing and the sports context. From this preliminary search in combination with the first author’s experience of the SFD literature, we created a search string that was also validated by a local university librarian. Specifically, early searches, and according to the first author’s understanding of this literature, sports interventions are by now an established field in itself, but the involvement of the police as an integral and explicit unit within the intervention is rare. Accordingly, we allowed for synonyms alluding to “police” to exist anywhere in the text and stringed this together with synonyms on “sports” and “youths.” The full search string, as copied from SCOPUS, is below.

(ALL (police OR “police officer” OR “law enforcement” OR “community policing” OR “police-community relations” OR “neighborhood policing” OR “proximity policing” OR “beat policing” OR “local policing” OR “democratic policing” OR “community-oriented policing” OR “reassurance policing”)

AND TITLE-ABS-KEY (“community sports” OR “recreation program” OR “youth sports” OR “sports participation” OR “physical activity programs” OR “sport” OR “sport associations” OR “sport club” OR “sport intervention” OR “sport program” OR “sport programme”)

AND TITLE-ABS-KEY (youth* OR teenager OR adolescents OR “young people” OR boys OR girls OR children OR juvenile OR “young adults”)

Subsequently, we searched six databases, ranging from more general and broad databases to more topic-specific databases. These were: Criminal Justice Database, EBSCOhost, PsycINFO, Scopus, SPORTDiscus and Web of Science. In accordance with Arksey and O’Malley’s (2005) guidelines, we also undertook a comprehensive grey-literature search. While there exist no established guidelines as to how to search grey literature, we took inspiration from several scoping reviews that concentrated on grey literature (Ubels et al., 2022). Such strategies included searching the web for relevant reports, utilizing parts of our search string, researching relevant organizations’ websites for any produced reports on the subject matter and contacting relevant practitioners for further additional information on the whereabouts of potential material. Back and forward searches were also performed on all final included articles to ensure an exhaustive search. The final search was conducted in April 2024.

We took inspiration from the Setting, Perspective, Intervention, Comparison and Evaluation (SPICE) model to properly structure the review and the study selection. The setting (S) was sports contexts (programs, initiatives, associations and interventions), perspective (P) alluded to theories or models on crime prevention for at-risk youths, intervention (I) pertained to the involvement of police officers in the aforementioned interventions and evaluation (E) pertained to the results emanating from such interventions on the relationship between youths and police. While we were mainly interested in relationship-building and trust as the outcome variable, we were also attentive to any variables that were affected by trust building between police officers and at-risk youth. However, trust and relationships, central to our review, had to be in the paper. Notably, most studies did not center on this particular variable but still contained findings on this relationship, often in conjunction with other variables of interest (e.g. crime reduction). These studies were included, although trust seemed to be accounted for as a mediating variable and not the direct outcome itself. However, if crime prevention or reduction was the only assessed variable without any findings on trust or relationships, the study was excluded. Examples of this include Hartmann and Depro's (2006) analysis of Midnight Basketball, which includes all relevant components but lacks any focus on trust, legitimacy, etc. Comparison (C) was deemed a non-applicable condition to apply to our scoping review. No time limit was imposed, and we allowed for peer-reviewed and grey-material to be included.

According to our SPICE model, in the first stage, abstracts had to mention some sort of (1) sport context, (2) contain at-risk youth or underserved areas and (3) outcome variables that directly pertained to police–youth contact and relationships or were relevant to a crime prevention purpose (e.g. crime frequency, recidivism or more broadly, antisocial behavior, development of pro-social behavior and more). Notably, given our pre-understanding of this literature, we allowed for the absence of synonyms to the police in the abstract-selection stage, since the police work in this body of scholarship is not always the center of attention, yet present to some extent in some studies.

After retrieving our full search, the three first authors analyzed 40 abstracts taken at random and rated according to the stated inclusion and exclusion criteria to determine interraterreliability. Given that we were more than two raters in this procedure, we calculated Fleiss Kappa, a reliability analysis suited to three or more raters. The procedure yielded a Kappa statistic of K = 0.919, indicating nearly perfect agreement. The first author then went on to scan the remaining abstracts, with the second author scanning a subset of abstracts (n = 100) to ensure continued interrater reliability. One single disagreement was found and discussed in this stage. After abstract scanning, we read the full-text articles. Notably, the articles had to be explicitly, and at least partially, concerned with police involvement with sport initiatives at the full-text stage. Some articles mentioned collaborations between a range of institutions and sport initiatives, such as the social services, the municipal and the police. However, if the nature of the cooperation between the police and the sports initiative was not further elaborated, the articles were excluded. Since this review aimed to map the effects of these sport programs and what factors moderated the effects, there needed to be empirical data on the cooperation and the relationship between police and youth, and not mere conceptual arguments or hidden assumptions. A range of articles made mention of interventions known to have cooperation with police departments (among others) but never discussed the role of police within these cooperations. Accordingly, these articles were also excluded, since no tangible outcome or measure could be deduced from these articles. The process is illustrated in the below flow chart. In total, ten articles met the inclusion criteria. Brake and Misener’s (2020) study was the result of Brake’s main thesis, and since the latter did not contribute any significant new information, only the peer-reviewed article was included. In contrast, Rabois and Haagan (2002) published a study from Rabois’s (2000) doctoral dissertation, but the latter contained additional experiments, hence its inclusion. The full study selection procedure is illustrated in Figure 1.

In the data charting process, we took inspiration from Peters et al. (2015) and structured our process according to their suggestion. An Excel sheet was created, containing a modified version of Peters et al.’s checklist. The sheet asked each scoping reviewer to list author information, year of publication, source of origin, aims/purpose, study population, methodology, intervention type, concept, duration of the intervention, how outcomes were measured and key findings relating to the review question. The sheet was first piloted to ensure it captured all relevant parameters of the papers.

The results section is structured in three parts. The first is a descriptive part, providing an overview of the included studies. In the second part, we address RQ2 by providing an overview of how sport interventions are supposed to reach their goal, thus illuminating the theoretical ideas and models behind the studies, after which we present the actual effects of these sport interventions. In the second part, we present the results relevant to program delivery from the police perspective, i.e. the factors that moderate contact with youths within the sport context.

Most studies were of a qualitative character (n = 5) and a subset were of either a quantitative (n = 3) or mixed-methods character (n = 2). The studies were conducted in the US (n = 5), UK (n = 2), Canada (n = 1), New Zealand (n = 1) and Sweden (n = 1), and most often within the context of the US-based Police Athletic League or similar initiatives, followed by grassroot sports clubs’ interventions. A full overview is provided in Table 1.

As shown in Table 1, there is a rather homogenous theoretical school-of-thought that underpins the included sport interventions. The main raison d’etre cited within the studies for conducting their intervention is, indeed, for youths to meet police officers under circumstances conducive to promoting healthy relationships, and much like other literature on police-youth relationships, a central theoretical concept here (Brake and Misener, 2020; Frazier, 2016; Broaddus et al., 2013; Rabois, 2000; Rabois and Hagaan, 2002; Treskon et al., 2020) is Allport’s (1954) contact theory. Contact theory posits that four conditions need to be addressed in order for interventions between divided groups to be successful. Firstly, participants should have equal status, where neither group holds a position of superiority over the other. Secondly, there should be common goals and meaningful objectives shared by the participants. Thirdly, to achieve these common goals, cooperation is needed. Fourth and finally, there should be institutional support; a healthy framework surrounding the activity that supports the desired behavior and intended outcomes within the intervention. Accordingly, it is based on this concept that a range of experimental and case studies are conducted (Brake and Misener, 2020; Frazier, 2016; Broaddus et al., 2013; Rabois, 2000; Rabois and Hagaan, 2002; Treskon et al., 2020), where sports are advocated as a platform especially conducive to the theoretical tenets of Allport’s theory. As seen in these studies, the sporting context can, theoretically, provide unique benefits in relation to Allport’s four conditions. Sporting competitions are, per se, considered a context where participants meet on equal terms and where social backgrounds should, ideally, be of little importance. Secondly, winning the competition is considered a main common goal within the team, thus providing a clear and meaningful objective of the intervention. Thirdly, the team naturally cooperates to achieve this intended goal and should foster mutual understanding and respect for each other’s efforts. Finally, the sporting context in itself is supervised by an institutional framework that emphasizes respect for rules, peers, competitors and which imposes rules and regulations to promote these factors.

As for the other included studies, Robbins (1990) report is rather descriptive and theoretical when considering the relationship between police and youth but cites the overarching COPS philosophy. Within this philosophy, as is commonly known to COPS scholars, police officers seek to establish mutual trust and confidence with their local surroundings. In this case, this includes a range of voluntary sports clubs in the UK context. Likewise, Höglund and Bruhn (2022) explore more broadly social integration and sports in underserved areas with a social capital theoretical lens (Putnam, 2000) and how local police officers constitute one important component in this equation. Social capital here is built upon the premises of close and personal contact, enabling youths to recognize police officers and other authorities and accumulate trust in these institutions. In Gordon et al.’s (2022) study, Bronfenbrenner’s (1979) ecological framework in conjunction with theoretical foundation from positive-youth-development (PYD) was used, mainly to understand how youths within the intervention developed important personal skills. Bronfenbrenner’s (1979) model is exhaustive and acknowledges the complex interplay between various levels of which the individual is embedded in, and one of these (micro)levels includes the interaction with local police officers. Similar to the other studies, the sports club provided a context for close and personal contact. Jeanes et al. (2019) utilized the concept of “boundary-spanning” – a theoretical model that centers on important actors that are able to facilitate complex and often multi-sectoral cooperations. The cooperation in question here refers to the sport coach and important institutions, therein the police. To become a relevant boundary-spanner, Jeanes et al. (2019) outline several critical factors. For example, the coach needs to be familiar with the local surroundings and have earned the trust of the local youths in the first place. Secondly, the coach needs to be well-connected to other institutions to facilitate the actual collaboration. When such factors are in place, the coach can “vouch” for, for instance, police officers to partake in the sport intervention with the other youths.

This section is structured according to two themes. The first is on the main outcome variable (trust), and the second is on an interrelated outcome variable, namely whether acquired positive views on police are generalized to the broader police population.

The effects on the police–youth relationship due to the sporting interventions are positive, with the exception of one study with non-significant findings (Frazier, 2016). In all other studies (Brake and Misener, 2020; Gordon et al., 2022; Broaddus et al., 2013; Jeanes et al., 2019; Rabois, 2000; Rabois and Hagaan, 2002; Robbins, 1990; Treskon et al., 2020), respondents frame their experiences as positive and explain how their initial suspicion and hostility towards police officers fade away as they get to know them through sports, or generally, how the sporting environment enables police officers and youths to engage with each other in a non-threatening and easygoing way (Gordon et al., 2022).

Starting with the qualitative case studies, key stakeholders in Höglund and Bruhn's (2022, p. 40) study argue that the contact is up-close and personal for police officers who come to partake in sports activities with youths: “When the youths see the community police officers, they don’t even see the police but instead it’s like ‘here comes *Name*’ and ‘there is *Name*’. It is a personal contact.” In Jeanes et al. (2019) study, the authors state that youths “… reported that Jane and Asmad [police officers] ‘were all right’, ‘a bit of a laugh you know … ’” and how “Jane and Asmad leveraged their relationships to prompt discussions about key problems or issues facing young people” (p. 441). These findings sit well with the other case studies. For instance, Brake and Misener (2020) describe how youths eventually enjoyed sporting with their local police officers, Broaddus et al.'s (2013) youths recognized the intervention’s police officers as decent and trustworthy, Gordon et al.’s (2022) police officers came to recognize the sport club as a milieu in which healthy relationships (with youths) were facilitated, and Treskon et al.'s (2020) references more broadly how youths developed more positive attitudes toward police. Accordingly, these quotes seemingly reflect stakeholders, police officers and youths’ views of the sport intervention as successful in facilitating positive contact between youths and police.

Complementing these qualitative case studies are the quantitative studies (Frazier, 2016; Rabois, 2000; Rabois and Hagaan, 2002). Competitive basketball teams in Rabois (2000) and Rabois and Hagaan (2002) were mixed of police officers and youth, and results indicated that youths generally conceived of their teams as cooperative and successful, and outgroup attitudes toward team members improved as a function of the intervention. In the follow-up study, these teams were rearranged, with the hypothesis that different types of contact with the police officers would moderate attitudinal change. Teams were arranged so that one group had a “no contact” condition, contact with one out-group exemplar and contact with multiple out-group exemplars (i.e. increasingly more police officers on the team). Similar to the first study, groups that contained at least one police officer found their team to be cooperative and successful. However, there was no evidence that increased contact (i.e. more police officers on the team) increased youths’ perception and attitudes of the intervention’s police officers. Similarly, police officer attitudes were also enhanced, indicating more positive views of the youths they met in the sport intervention (Rabois, 2000; Rabois and Hagaan, 2002). The only study showing non-significant results was that of Frazier’s (2016) volleyball intervention, where improvements in attitudes never reached statistical significance, although pointing to better attitudes. As Frazier (2016) himself points out, it is likely that the (very) limited sample size made it difficult to detect any tangible effect.

In relation to the previous section, some of the most cited mechanisms for these positive outcomes are directly or indirectly related to Allport’s theoretical assumptions about connecting diverse groups under certain circumstances. Meeting police officers in the sporting context is described as an opportunity to reframe police–youth relationship and show youths that police officers are regular humans too (Brake and Misener, 2020; Broaddus et al., 2013; Robbins, 1990). This is best done under conditions where neither party has a significant power advantage, such as in a sporting context.

Although the findings are predominantly positive in terms of youths’ attitudes toward police, a few studies (Broaddus et al., 2013; Rabois, 2000; Rabois and Hagaan, 2002) made note that the youths’ attitudes toward the interventions’ police officers rarely generalize to the broader police population.

From Broaddus et al.'s (2013) case study, they conclude that it is unclear whether these views are generalized to the broader police population. This is substantiated by experimental evidence. Results from Rabois and colleagues (Rabois, 2000; Rabois and Haagan, 2002) also pointed to deficits in generalizing positive attitudes. While youths in this sport intervention did appreciate police officers more after the completed intervention, the analysis also showed that the attitudes were specific to the very police officers who partook in the intervention. As Rabois (2000) states, it seems as if these youths instead considered the police officers in the intervention as not representative of the population, but instead a deviation. However, although youths’ attitudes in Rabois (2000) and Rabois and Hagaan (2002) were not generalizable, the police officers’ attitudes were. Police officers’ attitudes were enhanced toward the broader target population, indicating that the intervention had a more significant effect on police officer attitude than on youth attitude.

In terms of whether generalization is at all applicable, both authors and respondents in Broaddus et al. (2013), Rabois (2000) and Rabois and Hagaan (2002), in different ways, discuss whether their sample is, in fact, representative of the population they seek to understand. For instance, while some youths in Broaddus et al. (2013) gained a nuanced picture of police officers at the general level, some also remarked that they believed the police officers who volunteered for the sport intervention most likely were the ones who enjoyed spending time with youths. Such a remark highlights a potential scientific flaw, in that most of these studies, indeed, risk suffering from potential selection bias concerning the police officer samples. Similarly, Rabois and Hagaan (2002, p. 190) argue that the “… voluntary feature of intervention programs creates a potential difficulty in targeting hostile group members because those who participate may actually be the least hostile.”

It is clearly evident that, in at least some studies, the police officers in question are likely not representative of their population. For instance, in Robbins (1990) report on UK sport interventions, Robbins (1990) describes some police officers who voluntarily, and at their own expense, bring troubled youths with them to sporting competitions and such, and who also spend their free time on such activities. Similar cases are found in Brake and Misener’s (2020) study, with coaches go “… above and beyond …” their regular roles as police officers to help youths. While being an extreme case, such findings warrant caution about whether samples in these interventions are indeed generalizable.

In answering the second research question, we turn to how these sporting interventions are constructed, meaning the central component of this section is about program design. Here, we look at the program design from the paper’s unique point-of-view, i.e. the police perspective. Accordingly, we do not engage with the more generic features of program design in sport (see, e.g. Coalter, 2013) but maintain our focus on the police. Here, we carve out three relevant themes. These include how the police initiate and sustain contact with critical stakeholders, what role the police officers themselves play within the sport intervention and how this affects delivery and outcomes and how engaging with sport interventions is also a question of resource allocation.

The perhaps most significant finding in terms of factors that moderate program delivery is the contact the police have with coaches and important stakeholders of the intervention. In several studies (Brake and Misener, 2020; Gordon et al., 2022; Broaddus et al., 2013; Jeanes et al., 2019), it is evident that police officers who collaborate with sport interventions enjoy many benefits from having solid contact with the coaches therein.

For instance, Brake and Misener (2020) explained that, without the coaches as intermediaries who had already established report with the youths, these youths would probably not have engaged in the sport intervention. Similarly, Gordon et al. (2022) point out how the positive milieu in the sport club environment, mainly created by the existing coaches, is a perfect opportunity for officers to engage in and show youths what a healthy climate looks like and how relationships here look like. The most telling example is provided by Jeanes et al.’s (2019) study on sport coaches as trusted “boundary spanners.” The following quote illustrates the way the coach serves a critical role in the relationship-making between youths and police, starting with discussing police attendance with the youths:

In other studies, coaches could not only become key brokers in this relationship but also intervene as they spotted issues in the intervention. For instance, facilitators within Broaddus et al.'s (2013) intervention could correct police officers who overstepped boundaries or who seemed uninterested in the intervention, thus ensuring intervention quality.

The police role varies somewhat between studies, with some studies emphasizing the police officer role more than others. Specifically, in the actual interventions, police officers partook as participants/competitors, invited guests or as trainers and intervention leaders themselves.

In a range of studies, the police officer competes with, and against, youths (Frazier, 2016; Rabois, 2000; Rabois and Hagaan, 2002), and a significant question in particularly the studies conducted by Rabois is whether the constellation on particular teams affects relevant outcome variables. Other studies, primarily by Broaddus et al. (2013), emphasize a more collaborative approach, and in this case, through rock climbing. In this study, police officers are matched with youths and the rock-climbing activity becomes collaborative in nature. These kinds of conceptualizations of the police officer role generally build upon Allport’s idea of power relations and seek to minimize power hierarchies through establishing that everyone is competing as equals.

The other format of police involvement the studies revealed was interventions which were solely led by the police themselves, where police officers maintained a clear leader and trainer position. In one of Robbin’s (1990) studied interventions, a local boxing club was led by the police department. In many respects, this intervention resembled traditional sport interventions, where the focus is on youths’ socialization within the sport context. Little to no focus was given on the aforementioned importance of power dynamics between police officers and youths, but instead, the police officer acted as a strong leader and, proclaimed, role model. Relationships that grew here were more the result of a mentor–student relationship and did not build upon Allport’s (1954) idea of equal status. Interestingly, this particular intervention was accused by other social actors of potentially breaching youths’ trust and of being part of a surveillance scheme. Youths, in other studies, made mention of their suspicion of the intervention and feared “… that it was actually a set up” (Brake and Misener, 2020, p. 209) and that they risked being increasingly monitored. While police officers in Robbins’s (1990) study rejected the accusation of extending their surveillance, they nevertheless noted that their position enabled them to have information about ongoing events in the community that they could act upon.

The final theme here is that of resource allocation, and a range of studies discuss the need for police officers to argue with their departments for the importance of engaging with the sport interventions. As noted by Robbins (1990), a lot of the UK police officers engage voluntarily in sport interventions, which are inherently associated with extra work and discussions of how sustainable such practices are. This aspect is also tangible in Brake and Misener’s (2020) study, where the program director discussed in detail all the partnerships necessary for program delivery. One of the key partnerships here was with the police, forcing the community police officers to rationalize their involvement to their department and often dedicate some of their own time to the cause. The following quote from Brake and Misener (2020, pp. 9–10) is illustrative:

Time management and resource allocation are present in most works in this review and manifest themselves through different ways, not only in the empirical material. Rabois (2000) and Rabois and Haagan (2002) remark in a methodological discussion that their experiments and interventions were constructed so as to respect police officers’ already strained time and to honor their voluntary commitment. Finally, some results are worthwhile mentioning where priorities are not given to the sport interventions. For instance, in Treskon et al.'s (2020) study, stakeholders critically remark that police officers rarely showed up, and when they did show up, they rarely engaged in any meaningful interaction with youths. Accordingly, if the engagement with the intervention is of poor quality on the police officer’s behalf, this may be contraproductive to the intervention’s purpose.

In this review, we sought to map out the empirical literature on relationship-building between youths and police in the sport context, with a further aim to integrate police science and SFD literature. The overarching aim with this endeavor was to evaluate the effects of such interventions on youth attitude toward police and to understand the factors that moderate these attitudes. In this discussion, we discuss the results and the various facets of SFD and sport-based interventions in relation to community policing. Two areas dominate our discussion: (1) the review results in relation to the broader literature on community policing in low-trust areas and (2) the aspects of cautious programming on sport-based interventions to not misplace the focus of police roles in sport interventions.

We situate the first point of discussion in what Rukus et al. (2017) refer to as community policing’s greatest failure: that community policing is the least effective where it is the most needed, i.e. with low-trust groups and areas. In this regard, the review results are positive. The majority of review studies indicate that sport seems to be an appropriate arena for relationship-building between police and at-risk youth, and our discussion starts with the point of departure that this venue holds promise for relationship-building. Prior to engaging in the discussion in full, it is important to note that this evidence should be interpreted with care because of the sheer sample size of studies included here.

The main explanation for successful relationship-building in this context is frequently referred to as the role played by the coaches. A fundamental aspect of community policing refers to the co-production of safety with citizens and local community leaders (Skogan, 2006). Providing structured platforms for interaction between police and citizens, such as those of community meetings, can contribute toward “… building connections with hard-to-reach communities, and enhancing police accessibility” (O’Reilly, 2024). In such spaces, it is critical that police engage with key people who represent the community adequately and not only their own interests (MacColman and Dikenstein, 2023). Relatedly, the quality of leadership these representatives possess is of great importance when attempting to develop police–citizen relationships (Headley and Kalesnikaite, 2024). While we cannot say anything about the leadership quality of sport coaches as an aggregate group, they clearly work closely with the target group and can act as “brokers” and vouch for police officers to initiate trust-building processes with youths. To circumvent the poor perceptions of police that at-risk youths hold, this third-party broker seems conducive, and building this relationship could be integrated in police work.

To put this into a framework, we turn to community policing’s tactical dimension (Cordner, 1997), where problem-solving is an integral part (Reisig, 2010). In the context of this review, we suggest that community police officers who are looking to engage with at-risk youth establish contact and nurture relationships with local and influential sport coaches and initiatives as part of a wider problem-solving tactic. Preferably, these coaches’ expertise and local reputation should be properly utilized when designing sport interventions. Coalter (2013) and Holt et al. (2016) stress that the involvement of skilled and committed coaches can significantly enhance the effectiveness of sports interventions by providing structure, guidance and mentorship.

Another important consideration in structuring this tactical approach is the specific role police officers should assume in sports interventions. Research shows that perceptions of police actions and fair treatment significantly influence community cooperation and the legitimacy of law enforcement (Tyler and Fagan, 2008). However, negative experiences with police may weigh more heavily on individuals’ attitudes than positive ones, which could foster a lasting negative bias, especially among at-risk youth (Skogan, 2006). This is particularly critical for young people who frequently develop increasingly negative views of police during key formative periods in their lives (Sheeran et al., 2022). At the same time, research on police–youth engagement programs highlights their potential to positively influence youth perceptions of law enforcement, but that these programs require thoughtful design to maximize their impact (Azmy, 2021; Sheeran et al., 2022).

A cornerstone in this programming phase is to think about power relations. Allport’s (1954) contact theory, which underpins much of the research on police–youth interactions, suggests that structured, positive interpersonal contact can reduce prejudice and foster stronger relationships between groups. In practice, various studies on police–community interactions have used Allport’s framework to examine how structured programs impact public perceptions of law enforcement and find that meaningful interpersonal contact under equal-status conditions can reduce prejudice and foster positive relationships between groups (Azmy, 2021; Sheeran et al., 2022). Our second suggestion, supported by the findings of this review, is that police officers should avoid assuming authoritative roles in sports interventions. To promote an atmosphere of equality and foster genuine relationships with youth, police should engage as participants, collaborators or perhaps better, even allow youths to take the lead in such interventions, rather than leaders or figures of authority. By taking on roles that minimize hierarchical divides, officers can better support positive, trust-building interactions that align with Allport’s contact theory principles.

On a related note, it would seem important to structure this particular component with extra care. Drawing from findings in adjacent research on school-resource officers (SROs), discussions can be found where school spaces are conceived of as criminogenic hotspots (Higgins et al., 2022), risking the hyper-criminalization of trivial behavior with further police presence (Schlosser, 2014). This parallels the experiences in community sports, where police involvement can sometimes be perceived as an extension of social control rather than a genuine effort to build trust (Andrews and Bustad, 2017; Brake and Misener, 2020). Taking the case of the Police Athletic League in Baltimore, Andrews and Bustad (2017) argue that such initiatives embody a shift from public support for recreation to a model that positions recreation as part of the “social problems industry,” where sport is used as a vehicle for social control but without addressing root causes of social inequality and crime. This is a serious potential pitfall of sport interventions, since community policing, at its core, is meant to build trust with communities, foster partnerships and address the root causes of crime and social disorder (Skogan, 2006). Accordingly, there is a need to think carefully about sport interventions’ aims and make sure that trust-building is in the center and that notions of surveillance and governing are mitigated.

To avoid these pitfalls and foster genuine, trust-based interactions, sport programs can draw valuable lessons from critical reflections on SROs and similar initiatives. To reconnect with the previous point about the roles of police officers, one such lesson is that relationship building seems more appropriate when police officers take on roles as Allies and collaborators and less so when assuming roles with a focus on monitoring and discipline. Such roles can foster adversarial relationships, especially with young people.

While SFD literature has extensively explored the benefits of sports for youth development, it has not adequately addressed the unique contributions and challenges associated with police participation in these initiatives. Understanding the police’s role in these settings is crucial for developing effective strategies that foster positive relationships between law enforcement and at-risk youth. By stimulating interdisciplinary collaboration between police science and SFD, such reflections are perhaps more easily attainable. Finally, the review has certain limitations worth mentioning. Firstly, the review may underrepresent contextual differences across various SFD initiatives, such as regional or cultural variations, which can significantly affect how police involvement is perceived and implemented. Secondly, the focus on at-risk youth and underserved areas, particularly in the abstract and initial framing, may inadvertently overlook studies that explore community policing’s broader impact on building trust and legitimacy across a range of community settings. Lastly, the review does not include a substantial analysis of long-term impacts of police-involved SFD programs, leaving a gap in understanding the sustainability and effectiveness of these initiatives over time.

This paper was funded by an internal grant at Södertörn University, The Swedish National Council for Crime Prevention and Stieg Larsson's foundation.

Declaration of interest statement: We have no conflict of interest to declare.

Corrigendum: It has come to the attention of the publisher that the article, Tony Blomqvist Mickelsson, Anders Nordström, Chato Rasoal, Mehdi Ghazinour; Community sports as a conduit for police–youth relationships: a scoping review. Policing: An International Journal; 10.1108/PIJPSM-07-2024-0105 included the wrong reference. The reference should read, “Brake, J., & Misener, K. (2020). “It'sa ripple effect”: the role of intergroup contact within an inner-city youth sport for development and peace program. Managing Sport and Leisure, 25(3), 203-219”. The publisher asks that reference be entered correctly at submission and confirmed at article proofing stage.