Community-based participatory research on the impact of greenspace on violent crime

Background. The positive impact of greenspace on human health has been well documented, including several literature reviews and meta-analyses that have examined the broad benefits of nature connections. Researchers have also examined the relationship between nature and crime reduction and identified potential mechanisms underlying this outcome, such as the physiological impact of nature, lowered temperatures due to a reduction in the heat island effect, and places for community interaction. However, a critical shortcoming of this study is the lack of deep community involvement in the research process. Community-based participatory research (CBPR) is critical to ensuring that the findings are meaningful to communities and translatable. This study expands on recent literature reviews on greenspace outcomes by focusing on community-engaged research (CER). By gathering and summarizing studies on this topic, we address two subjects: (a) strategies that can be used to improve community engagement, and (b) environmental factors that impact community outcomes in greenspace settings. Methods. To explore these issues, we used a modified version of Arksey and O’Malley’s framework for a structured literature review, employing the Web of Science, EbscoHost, Scopus, ProQuest Global, and Google Scholar databases. Results. We retrieved 772 publications using permutations of keywords related to violent crime, greenspaces, and CBPR. After eliminating duplicates, the reviewers worked in parallel to evaluate 700 titles and abstracts and identified 51 potentially relevant papers, ten of which met the requirements for inclusion in this analysis. Discussion. Based on the studies explored in this literature review, we identified the following strategies for improving CER: building partnerships, facilitating power-sharing, utilizing community-specific indicators of success, embracing perspectives of communities of color, and empowering community researchers. In the sample of studies described here, the factors contributing to the relationship between greenspace and violent crime were maintenance, activity programming, green interventions, and community involvement.


Introduction
This structured review investigated the relationship between the impact of greenspace and violent crime in the context of community-engaged research (CER). By gathering and summarizing studies on this topic, we address two subjects: (a) strategies that can be used to improve community engagement, and (b) environmental factors that impact community outcomes in greenspace settings. In this paper, we refer to CER and community-based participatory research (CBPR). For our purposes, CER refers to any study in which researchers gather information regarding community circumstances or needs. On the other hand, CBPR suggests a deeper involvement in which community members take part in the definition of the goals and priorities of the project, ideally engaging them as members of the research team. CBPR and related participatory approaches, such as participatory action research (PAR) and action research, are characterized by a deep respect for people's ability to understand and address issues confronting their communities (Brydon-Miller et al 2003).
Research conducted in communities or using community residents rarely includes them in a participatory manner, resulting in a power asymmetry between researchers and community members (Lincoln 2001, Hacker 2013. By disregarding local standpoints and knowledge, researchers might be prone to unintended bias in shaping the questions asked and interpreting the results-for example, missing important issues regarding stakeholder experience or misinterpreting findings due to lack of contextual or cultural awareness. As a result, vulnerable and marginalized communities may not receive direct benefits from the research conducted on them and may be exploited by investigators (Smith 1999, Lincoln 2001, Hacker 2013. The goal of CBPR and related approaches is to produce practical knowledge relevant to the life circumstances of communities and the people who reside within them (Reason andBradbury 2008, Hacker 2013). Participatory approaches create translation processes and bidirectional connections between local and academic knowledge Levin 2003, Hacker 2013). Moreover, participatory research aims to enhance stakeholders' agency and their capacity to identify and address challenges. These findings serve as the foundation for processes associated with fostering social change (Cammarota and Fine 2008).
This study explored CER as a tool for investigating the mechanisms and design qualities that might result in safer and healthier neighborhoods. However, a critical shortcoming is the lack of deep community involvement in the research process. CBPR is essential to ensuring that quantitative results are meaningful to the community and representative of their lived experiences. In other words, CBPR aims to ensure that the voices and knowledge of those most affected by the intervention are listened to and valued.

Background
To create a context for discussing the findings of our structured literature review on greenspace, crime and community engagement, we needed to review the broader associated literature. This process was constrained by the lack of studies addressing all three variables simultaneously. Prior to summarizing the results of the primary literature review, we provide background on greenspace and crime reduction and greenspace and community engaged research (see figure 1). Shepley et al (2019) conducted a scoping review of papers and studies evaluating the relationship between nature and crime reduction in the United States and noted that some greenspaces were more productive than others. Studies on parks have varied, possibly because of the variety of features and level of maintenance. Community gardens, vegetated streets, trees, and ground cover were found to be effective. Undeveloped green areas did not appear to have any effect. The authors proposed potential mechanisms underlying these outcomes, such as the physiological impact of nature (e.g. lower cortisol levels), lowered temperatures due to a reduction in the heat island effect, and places for community interaction (pp 9-11).

Greenspace and crime reduction
Shepley et al noted the following impacts on crime based on type of greenspace: (a) The impact of parks was inconclusive, with three studies noting a reduction in crime, three showing an increase in crime, and four showing inconclusive results. (b) Community gardens appeared to be more effective, as six studies indicated a reduction in crime, one with an increase in crime, and five with inconclusive results. (c) The impact of vegetated streets and urban greenways was very promising, as four studies indicated a reduction in crime, no studies indicated the contrary, and one had no conclusive results. (d) Trees and groundcover were also very promising, with nine studies reporting positive outcomes and five inconclusive outcomes.
Updating Shepley et al (2019) reviewed and shifted to incorporating more international research, we found the following studies based on these four types of greenspaces. Polko and Kimic (2021), in a survey of park users in Poland, noted that the conditions of the equipment, park paths, and pavement were strong indicators of park safety. Potential criminals could interpret this in the same way and consider these conditions to be inhibiting the opportunity to commit a crime. In a study in Indonesia using data provided by urban wards (municipal administrative divisions), Sukartini et al (2021) found that an increase in greenspace resulted in less crime and a decrease in greenspace resulted in more crime. Several studies have been published on the impact of greening blighted vacant lots. The resulting green amenities can be interpreted as informal park settings. Moyer et al (2019) studied the impact of vacant lot remediation and confirmed the research of others that minor improvements in mowing and trash removal result in less gun violence. Similarly, Macdonald et al (2021) found that crime was reduced overall in remediated lots, even in neighborhoods close to train stations and retail alcohol establishments. The authors noted the important contribution of simultaneous measures in addition to physical environmental interventions to enhance social cohesion (Macdonald et al 2021). Sadatsafavi et al (2022) built on before-and-after studies to create a meta-analysis and benefit-cost study that confirmed the finding that remediation can contribute to a reduction in gun violence. While they concluded that remediation is less expensive than implementing law enforcement policies, it does not cover all program costs such as maintenance (Sadatsafavi et al 2022).

Community gardens
Recent case studies have suggested the effectiveness of community gardens (McCabe 2014). Beam et al (2020) used quasi-experimental methods to study the impact of converting vacant lots into community gardens by comparing garden lots with planter boxes, benches, and small landscaping projects with unimproved control lots. Crime data were gathered from the municipal database and compared to imagery from satellite and ground data, and municipal records (Beam et al 2020). While theft was not affected, violent crime diminished (Beam et al 2020). Focusing on community gardens' growth between 2005 and 2015, Koop-Monteiro (2021) gathered data on new community gardens, income, ethnic diversity, population, homeownership, and other demographic data and found that the addition of just one garden reduced property crimes (breaking and entering and vehicular theft).

Vegetated streets and urban greenways
The development of vegetated streets and urban greenways has increased in recent years (Horte and Eisenman 2020). Examples include the Highline in New York, 606 in Chicago, and the movement to convert railways to biking/walking trails, such as the Atlanta BeltLine Trail. In a literature review focusing on crime prevention and street characteristics (networks, paths, and nodes), Mao et al (2021) found that street network qualities such as distance, grade, and ambience impact crime. Path activities, visibility, and lighting were critical to deterrence, while nodes that provided the opportunity to escape or had poor visibility were associated with (Mao et al 2021).

Trees and groundcover
The possibility that trees and ground cover have a positive impact on crime reduction has been supported by recent studies. Schertz et al (2021) used cell phone location data to calibrate park visits and street activity. Both these independent variables (high frequency of park visits and high level of street activity) were associated with reductions in crime. Researchers have concluded that the confluence of tree cover, street activity, and park activity together contribute to reductions in crime (Schertz et al 2021).
Related to the presence of greenspaces, trees, and grass, Dong and Liu (2022) focused on assault outcomes and found that when walkability through the environment increased, the environment was safer. This is particularly true for grassy areas.

Community based participatory research
Participatory approaches, such as CBPR, PAR, and action research, involve equitable partnerships between researchers and community members (Lincoln and Guba 2000). Participants tend to be members of social groups who rarely participate in the academic production of knowledge, such as minorities and marginalized and underserved communities, so that decision-making that directly impacts their lives is taken from the standpoint of outsiders (Gaventa and Cornwall 2007). Participatory research frequently embodies a counter-hegemonic approach to knowledge production by collaborating with vulnerable and marginalized communities (Kindon et al 2007, Bradbury-Huang 2010. CER publications found in this portion of the preliminary literature review tended to focus on the most appropriate strategies or tools for gathering data. King et al (2021) discussed the geotagged OurVoice app as a tool for working internationally with youths to explore topics related to health inequities. This tool, which supports community-engaged participation and citizen science, allows participants to download photos and narratives and make observation-based evaluations. King et al (2021) used this tool to enable participants to make observations about environmental features that promote health behaviors or conditions and share them in group meetings. The comments of the youth/citizen scientists in this study varied but sometimes focused on environmental justice.
Participatory mapping is another tool employed by researchers to engage communities. Douglas et al (2020) gathered crime data and used a participatory geographic information system tool to evaluate neighborhood features related to public parks. The researchers found that tobacco shops were the foci for environmental disparities and argued that significant findings would have been lost if the methods were limited to surveys, thereby disenfranchising the community (Douglas et al 2020).
As mentioned, a foundation of studies exists around health, crime, and community engagement, but the intersection around greenspaces has not yet been fully developed. However, studies involving community members in CBPR are critical to ensuring that the findings are meaningful and translatable to meet the unique needs of individual communities. The following section describes research on the intersection of these topics and the focus of the present structured review.

Definitions
In this paper, the three primary keyword categories were 'green space' , 'violent crime' , and 'community-based participatory design' .
Greenspace. In the academic literature, greenspace or greenspace is defined as 'synonymous with nature' and 'explicitly urban vegetation' (Taylor and Hochuli 2017). Greenspace was defined using the broad description provided by the US Environmental Protection Agency (EPA) (2022). According to the EPA, green space is 'land that is partly or completely covered with grass, trees, shrubs, or other vegetation… Green space includes parks, community gardens, and cemeteries' .
Violent crime. In the Federal Bureau of Investigation's uniform crime reporting (UCR) program, violent crime comprises four offenses: murder and non-negligent manslaughter, forcible rape, robbery, and aggravated assault. Violent crimes are defined in the UCR program as 'those offenses which involve force or threat of force' (Federal Bureau of Investigation (FBI) 2018).
Community-based participatory research (CBPR). CBPR describes a collaborative approach to research that equitably involves all partners, including communities, in the research process, and recognizes the unique strengths that each partner brings to the process (Minkler and Wallerstein 2003). Beginning with a research topic that is of interest to community members, CBPR aims to bridge the gap between knowledge and action to support social change initiatives that 'improve community health and eliminate health disparities' (Minkler and Wallerstein 2003, p 4). We considered CBPR studies with at least one of the following three features: (a) engaging community members as members of the research team; (b) engaging community members in the implementation of the intervention and activities; and (c) counter-hegemonic character-that is, obtaining an in-depth understanding of the perceptions, priorities, and lived experiences of minorities as well as efforts to address inequities in power between researchers and community members.
We used a modified version of Arksey and O'Malley's (2005) framework to conduct a structured literature review, which included (a) identifying the research question, (b) identifying relevant studies, (c) selecting studies, and (d) collating, summarizing, and reporting the results. Articles written in English, Spanish, Portuguese, and French from 1970 onward that covered research in urban areas (not rural or suburban) were eligible for inclusion. The independent variable had to include at least one type of greenspace. The methodology must include some form of CBPR. At least one of the dependent variables had to be violent crime or perceptions of violent crime. The search focused on original, primary peer-reviewed literature, although doctoral dissertations, white papers, and conference proceedings were also eligible. Conflicting opinions regarding the relevance of a particular paper were resolved by a third independent reviewer. The independent reviewer was blinded to who had made the evaluation and the motivations for making the decisions, thus enhancing the validity of the evaluations.

Search strategy and database selection
The databases were selected based on their relevance and journal coverage. Subject area-specific databases were identified, and a broad interdisciplinary database (Scopus) was searched. Initial searches were conducted in December 2021, with updates to the searches run in February 2022.
• Using the EBSCOhost research platform, a joint search was performed using PsycINFO and CINHAL.
• Using the Web of Science research platform, the following indices were searched: Core Collection, CAB Abstracts, and SciELO. • ProQuest Dissertations and Theses-Global was searched using ProQuest.
• PubMed was searched using MeSH terms.
• To ensure relevant grey literature was considered, Google Scholar was also searched.
The searches were de-duplicated using Mendeley reference software and then uploaded to the Covidence software for literature reviews, where another round of de-duplication took place. After the de-duplication efforts were completed, two researchers conducted a blinded title and abstract screening in Covidence. The conflicts were resolved by a third researcher. Full-text articles emerging from the screening process were evaluated by two researchers and eliminated if they did not meet the inclusion criteria. Conflicts were resolved by a third blinded researcher.
A list of search terms for each of the three search categories was drafted using keywords and terms from papers that were known to be relevant to the review. Following testing in the chosen databases, the following terms were used in all searches, although the syntax of the search was adapted per database requirements, as necessary: (garden * OR green area * OR green-place * OR green-space * OR greenery OR greening OR greenness OR greenspace * OR greenway * OR natur * OR 'outdoor environment * ' OR park * OR trail * OR tree-canop * OR 'tree cover' OR urban-forest * OR 'urban design' OR 'urban forest * ' OR 'urban green * ' OR 'urban nature' OR vegetation) AND ('armed robber * ' OR aggression * OR assault * OR crim * OR 'crime prevention' OR crime statistics OR criminology OR firearm * OR gang OR gun OR guns OR homicide * OR manslaughter * OR murder * OR 'public safety' OR rape * OR violen * OR 'violence prevention') AND ('action research' OR CBPR OR 'collaborative research' OR 'community engagement' OR 'community participation' OR 'community-based participatory research' OR 'community-based research' OR 'implementation science' OR 'participatory research').

Evaluation process
Two researchers independently screened the titles and abstracts. A third researcher resolved the conflicting decisions. Once the final list of papers was established, it was reviewed by the first two researchers to confirm that they were appropriate for inclusion. As with previous rounds, when the two reviewers disagreed on their evaluation, the third reviewer resolved conflicting decisions.
Once the final list of papers was determined, one researcher entered the summary information into the literature matrix (see figure 2).

PRISMA summary
In February 2022, 772 records were identified through a database search. After excluding duplicates, 700 articles were screened. After subsequent title and abstract screening, 51 articles were selected for full-text evaluation. Ultimately, ten papers were selected for inclusion. These papers were the only ones from the group that met the criteria for addressing community participatory research, greenspace, and violent crimes (see table 1).
The resulting collection of papers falls into two categories. The first category (section 4.2) included articles that addressed exceptionally high engagement of community members as research participants. While not serving as part of the research team, the interactive process resulted in content that would enable the support of future research. Although the original intention of the authors was only to address full research team engagement, the literature was so sparse that we decided to fold in the handful of studies that suggested a more thorough involvement of community members. The second category of papers engages community members as part of the research team (section 4.3). This approach is the most valuable, as it allows community members to participate in the development of tools as well as the study. The ten papers (summarized in table 1) were allocated into these two categories, although one (Hite et al 2017) was listed twice as it involved two parts.

Engagement as research participant 4.2.1. Interviews, focus groups and photovoice
Despite similar design characteristics and maintenance programs, the impact of greenspaces may vary across communities. Harris et al (2020) used qualitative and quantitative methods to study the impact of The 606, a greenway intended as part of a revitalization effort in Chicago, connecting neighborhoods of differing racial/ethnic/socioeconomic compositions. Researchers observed the users of The 606 by employing the system for observing play and recreation in communities (SOPARCs) tool and conducted structured interviews (Harris et al 2020). The principles of community engagement aimed at preventing power inequities were addressed by working with community leaders to help build trust with community members (Harris et al 2020). Recording devices were not employed to help reduce the appearance of a hierarchical relationship between the researcher and participants, and participants were also given the opportunity to review the quotations. Some outcomes from the study indicated that 606 was perceived as providing a 'safe space' for community interaction, as local parks prior to 606 were associated with the presence of local gangs, but concerns were also expressed regarding gentrification and exclusion (Harris et al 2020). Differences in use were noted between groups: the Latinx population was more likely to use it for social purposes, while the White population used it for exercise (Harris et al 2020).
The Latinx population was also a population of interest in research by Stodolska et al (2009), who used focus groups to obtain feedback from communities on the impact of perceived crime on recreational behavior in their neighborhoods' natural environments, such as parks. The conversation also touched on the perceived seriousness of crime in their neighborhoods and whether safety was of concern when visiting parks Youth and adults were engaged in a photovoice activity to support collective efficacy as part of a larger crime prevention project.
Process indicated potential strategies for intervening with broader neighborhood problems, while also facilitating stronger social ties between neighborhood residents.
(Continued.) The redesigned space reduced the presence of criminal activity in the area, demonstrating that changes to the community-led physical environment changes can reduce illegal activity and may be a useful prevention strategy. (Stodolska et al 2009). Participants acknowledged that gangs were a common presence in the parks and noted that this was a common reason they did not frequent those spaces (Stodolska et al 2009). Ultimately, this fear drove avoidance behaviors, protective behaviors (e.g. relying on others), or collective behaviors (cleaning up parks) (Stodolska et al 2009). Participants also noticed that poorly maintained parks attract gang activities and suggested that cleaning up the parks and providing recreational activities to local youth could help break the cycle of violence. In a study of alleys in Los Angeles, California, Wolch et al (2010) conducted spatial analyses, physical environmental audits, focus groups, and behavioral observations. Four community organizations were involved: TreePeople, Trust for Public Land, Pacoima Beautiful, and LA Neighborhood Land Trust (Wolch et al 2010). The participation of community groups was limited to focus groups that addressed adult and adolescent perceptions of alleys, how they might be used, and the memories they generate (Wolch et al 2010).

Community activities and events
Working within the context of three community programs-crime prevention through environmental design (CPTED), attention to youth at risk (AYAR), and prevention of gender based violence (PGVB)-Montemayor (2019) describes a research project intended to identify environmental infrastructure to support violence reduction (Montemayor 2019). Among the principles associated with CPTED is community participation in the development and maintenance of the neighborhood's environmental features. Given the high risk of young men becoming either victims or perpetrators of violence, the recruitment of at-risk youth through an AYAR approach was considered critical for the successful implementation of this research into practice (Montemayor 2019). Women subjected to violence are also engaged in the implementation process (Montemayor 2019). Three primary social diagnostic tools were used: community walks, drawing sessions with children, and community-mapping events that took place in open public spaces and engaged community members (Montemayor 2019). A critical part of the process involved the creation of neighborhood groups, accomplished in collaboration with the local government (Montemayor 2019). The outcomes varied across locations. Importantly, researchers have noted the complexity of this approach and the need to establish manageable geographic definitions (Montemayor 2019).

Communities as implementation forces
Another way in which communities are engaged in the research process is through their participation in the physical implementation of independent environmental variables, such as participating in the remediation of vacant lots or planting and maintaining community gardens. While most efforts to develop outdoor greenspaces are led by professional landscapers, Heinze et al (2018) were interested in whether participation of community members in the design and development of greenspace interventions would have a more positive impact on crime reduction than greenspaces developed without their involvement. A local program in Flint, MI solicited Grant requests from communities to mow, weed, and remove trash from vacant lots (Heinze et al 2018). Data were gathered on neighborhood disadvantage factors, resident attitudes, prior crime incidents, and the ratio of homes to vacant lots (Heinze et al 2018). Heinze et al (2018) concluded that there was a 40% reduction in crime in street segments in resident-maintained programs as opposed to those without maintenance.
Similarly, Hite et al (2017) described an ethnographic CBPR study conducted with community members in Frenchtown, a historically African American neighborhood in Tallahassee, Florida. In response to the uneven geographic development and food insecurity experienced in this community, residents have engaged in a local effort to reclaim and transform community spaces through the development of community gardens (Hite et al 2017). The study revealed that community members' direct involvement in the development and maintenance of community gardens promoted community health, maintenance of culture and heritage, placemaking, and economic opportunities, while also representing a powerful form of social resistance to structural racism and uneven opportunities (Hite et al 2017). Hausman et al (2009) engaged community members via focus groups to identify indicators of violence, after which the researchers followed up to see if data on these topics were available. Members from the Philadelphia Area Research Community Coalition helped create the focus group format, and other members participated in note-taking and the analysis of the information that was gathered (Hausman et al 2009). Data were also gathered from a horticultural society regarding parks, as defined by the percentage of tree canopies (Hausman et al 2009). The authors recommend that funding be provided to support community participants to reduce the burden of their participation (Hausman et al 2009).

Engagement as researcher 4.3.1. Community members as creators of research tools
In the early stages of their ethnographic study on the impact of community gardens in Frenchtown, Hite et al (2017) extended an open invitation to daily meetings with local partners in the community at large. Ultimately, this early engagement with the community led the researchers to frame the focus of their study on (a) the role that community gardens play in transcending race and the structural barriers of uneven geographical development, and (b) the availability of critical resources (Hite et al 2017). Research questions and methods were also developed in partnership with community partners and residents (Hite et al 2017). The methods identified through this process included photovoice, participant observation, interviews, free lists (a process in which participants list all words associated with a topic), and social network analysis (the study of relationships between individuals).

Community members as gatherers and analyzers of data
Photovoice has become a commonly used method to actively engage community members in research. Ohmer and Owens (2013) contacted local resident organizations and stakeholders and conducted informational meetings to gather feedback to formulate a photovoice project on crime prevention. Community members photographed the environmental features that they believed to deter or contribute to crime, and subsequently analyzed and prioritized the photographs. Adolescent community members developed presentations and shared project ideas, eventually expressing a strong interest in developing a community garden or art project (Ohmer and Owens 2013). Researchers note that identifying issues of concern and building relationships between community members supports collective efficacy as defined by mutual trust, social ties, and willingness to intervene in neighborhood problems (Ohmer and Owens 2013). The authors emphasize the need to establish a strong partnership with community-based organizations and devote extensive time to cultivating these relationships on-site (Ohmer and Owens 2013).
Peréa et al (2019) engaged urban youth as researchers by assigning the tasks of assessing local parks and proposing evidence-based health initiatives. To recruit six young researchers, applications were shared with the local youth task force, schools, and a job organization (Peréa et al 2019). Responsibilities include supporting the implementation of the assessment, fieldwork protocols, analysis of data, dissemination, and promotion of the program (Peréa et al 2019). Training involved exposure to issues regarding healthy cities and social determinants of health, as well as in the use of the system for observing physical activity and recreation in communities (SOPARCs) and the physical activity resource assessment tools (Peréa et al 2019). In addition to gathering data, youth researchers were interviewed about their personal experiences (Peréa et al 2019). The health promotion strategies generated by the youth included a newsletter, newspaper articles, and a report on the park system, although the ultimate decision-making power resided with the principal investigators (Peréa et al 2019).
A case study by Vidal et al (2021) involved collecting qualitative data during community meetings in anticipation of an environmental intervention that promoted play and social interaction in a neighborhood with high rates of violence. The meeting was facilitated by a neighborhood design center with the researchers serving as scribes, while the residents discussed the environmental conditions and use of public spaces and open lots (Vidal et al 2021). A Baltimore leadership group conducted a door-to-door survey on solutions to community violence (Vidal et al 2021). Two concerns expressed by community members include the high exposure of children to trauma and limited opportunities for positive experiences, as well as the fact that crime can inhibit people from walking and playing outside (Vidal et al 2021).

State of the research: CBPR and greenspace interventions
CBPR on the impact of greenspace is limited. This literature review found only ten papers that met the criteria for addressing CBPR involving greenspace and violent crime. The authors of these papers argue that more participatory research is needed on community-level interventions to transform urban environments. An important factor contributing to this scarcity is the time and effort necessary to conduct this type of study: forming strong partnerships, building trust among community members, and cultivating relationships, which is not always valued by the academic establishment.

Perspectives of communities of color
The use of CBPR on the impact of greenspaces on violent crime can highlight the perspectives of people of color. Absence of greenspaces in many racialized urban areas in US cities has not been accidental, but the result of historical discrimination and racial residential segregation (Wilson 2020, Nardone et al 2021, Huang and Sehgal 2022, Noelke et al 2022. A critical shortcoming in the research to date has been the lack of community involvement in the research process, disregarding the perspectives and lived experiences of African American and Latinx communities concerning greenspace interventions. In other words, studies on the benefits of greenspaces have replicated the same exclusion that has marked city planning and design in US cities.
A few papers in this study have explicitly addressed the lived experiences of people of color and the different meanings they attribute to their experiences with greenspaces. Stodolska et al (2009), Peréa et al (2019, and Harris et al (2020) focused on the experiences of Latinx communities, while Hite et al (2017) focused on an African American neighborhood. They suggest racialized communities may interact with greenspaces in particular ways, due to both cultural practices and persistent social oppression. Other studies have characterized their partner communities as low-income, disadvantaged, vulnerable, or disinvested, without specifying the particular cultural background and social dynamics of racialized communities. Arguably, a proper understanding of cultural specificities and power dynamics might contribute to the acceptance and success of community-level interventions in the built environment. Harris et al (2020) indicated that white and Latinx populations use greenspaces differently: White users are more likely to engage in the space for vigorous exercise (active recreation), whereas Latinx communities often use it to gather and socialize (passive recreation). Nonetheless, for Harris et al, communities of color have little say in the design and development of green infrastructure. This disregard for cultural appropriateness is particularly troublesome as enhanced community interaction and use of greenspaces for socialization have been identified as factors that may help prevent crime (Ohmer and Owens 2013, Hite et al 2017, Vidal et al 2021. Therefore, we recommend exploring and acknowledging different perspective and cultural behaviors regarding the use of greenspaces.

Creating partnerships
The papers included in this study describe the long process of creating partnerships with reputable community-based organizations and building trust with community members. Most papers discuss the amount of time and activities needed to get to know the community and create meaningful relationships before developing research methods and gathering data. This initial phase was essential to enable the community to take part in defining the priorities of the project and framing the research questions. Sharing control over the project will ensure that the community benefits from work.

Power sharing
CBPR projects strive to address the uneven dynamics involved in research projects, where investigators hold more power and privilege than community members, with the goal of creating equitable partnerships. The first step towards equity is avoiding stereotyping researchers as organizers, leaders, and community members as participants and followers, and recognizing the intelligence and capacity of community groups to understand their circumstances and take action. Most papers included in this study described how they addressed power asymmetries at play, communicated with community members, and shared ownership and decision-making power. However, none of the projects were perfect. Transparency regarding the limitations and institutional constraints surrounding the project is fundamental to maintaining mutual respect and trust (Jahanzoosi 2006).

Community-specific indicators of success
CBPR studies enable the emergence of new questions and success indicators identified by the community, thereby increasing local relevance. Hausman et al (2009) stressed the importance of understanding a community's vision of success as a touch point for collaboration between academic and community partners. Their study focused on translating the community-identified indicators of effective violence prevention into measurable variables. Similar processes might be time-consuming but are critical to understanding the mechanisms and design qualities that might result in safer neighborhood experiences in specific cultural and social contexts.

Empowering research
Engaging community members in the research team might enable not only a better understanding of context-specific factors but also enhance the community's capacity to take action and collective efficacy. One of the main criticisms of conventional research on vulnerable and marginalized communities is that investigators gather, depart, and leave nothing behind (Hacker 2013). Ohmer and Owens (2013) suggested that engaging community members as gatherers and analyzers of data in a photovoice project facilitates connections among residents to identify and act on issues of community concern. Participatory methods such as photovoice can be used to facilitate intergenerational dialogue, build connections between community members, and enhance social ties that are fundamental to collective efficacy (Ohmer and Owens 2013). Hausman et al (2009) argue that participation in evaluation activities might support strong community involvement in the intervention, and Peréa et al (2019) suggest that engaging youth in the research process may lead them to develop valuable skills, such as leadership, public speaking, presentation, and advocacy skills, and increase their confidence and sense of agency.

Factors contributing to the impact of greenspace on violent crime
The factors discussed in these ten stakeholder-informed papers included maintenance, activity programming, green interventions, and community involvement. Stodolska et al (2009) and Harris et al (2020) suggest that the connection between parks and crime reduction is related to park maintenance. They proposed that park maintenance, creating beautiful and orderly spaces, might inhibit the presence of gang members.
Another factor mentioned is the importance of organized activities offered to the community, such as events, free exercise classes, and recreational activities for children (Stodolska et al 2009, Vidal et al 2021. Vidal et al (2021) argued that the implementation of activities could help a community reclaim a public space in a neighborhood with high levels of violence.
As previously mentioned, smaller types of greenspaces (i.e.street trees, greenways, or community gardens) have shown a promising impact on reducing crime. Vidal et al (2021) argue that while often only high-cost comprehensive interventions are considered as worthy of implementation in underserved communities, small changes that are driven by members of a neighborhood can positively enhance community relations and increase attachment to the space and pride in maintaining it. In other words, bottom-up and lower-cost initiatives might be as (or even more) effective than comprehensive top-down interventions.
Several studies (Hausman et al 2009, Ohmer and Owens 2013, Hite et al 2017, Heinze et al 2018, Vidal et al 2021 have highlighted the positive impact of involving the community in the implementation of interventions and maintenance of greenspaces as a crime reduction strategy. They suggested that community involvement increases social capital and collective efficacy. Community interventions in greenspace can promote playful social interactions that strengthen social ties and promote mental health (Vidal et al 2021). Participants, in Ohmer and Owens (2013), remarked that the willingness to work together to address neighborhood problems might unify and build connections between more people in the neighborhood, therefore enhancing mutual trust and social support. Burrage (2011) suggests that green interventions enhance the interconnectedness of neighborhoods. Heinze et al (2018) noted that community-led greening programs are less expensive than government-initiated programs and contribute to community health and well-being by supporting collaboration and a sense of local ownership. The participatory involvement of community members in both the research and implementation processes denotes the agency-focused character of CBPR, in which stakeholders are considered capable of operating for their own well-being.
Academic researchers should be aware that volunteering in such activities might burden communities that are already facing disinvestment and hardship. Renumeration for community members' participation in research and greening activities is a critical hurdle in implementing participatory research. Academic funding may cover researcher and student participation, but rarely provides funding for community participants. To create equitable partnerships and value local knowledge and assets, compensation must reflect the contributions of all.

Conclusion
Based on the studies explored in this literature review, we identified the following strategies for improving CER: building partnerships, facilitating power-sharing, utilizing community-specific indicators of success, embracing perspectives of communities of color, and empowering community researchers. In the sample of studies described here, the factors contributing to the relationship between greenspace and violent crime were maintenance, activity programming, green interventions, and community involvement.
This literature review was limited to publications accessible through commonly available databases. Other research may have been conducted by communities and organizations that are not available through library search systems. As a result, critical studies, may be overlooked. While peer-reviewed literature is considered to be the 'gold standard' , we recommend availing oneself of reports produced by community and municipal organizations.
Another limitation of this literature review on CBPR and greenspace is that it was reviewed by only one community member. We recognize the contradiction in writing a paper about participatory approaches without the participation of a range of stakeholders. This study benefited from the input of additional stakeholders in the interpretation of the results and discussion.
This literature review may serve as a foundation for future CBPR studies that focus on the impact of greenspace on community wellness. Enabling community members to play a significant role in research teams will inevitably result in a higher-quality evidence-based design.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary information files).