The project was developed through a collaboration between two clinical psychologists (PS and KM) with expertise in suicide and violence prevention, respectively. The impetus for the project and the adoption of a PAR approach was largely driven (1) by KM’s existing community-based work in violence prevention, in which the urgent need to address firearm injuries and fatalities was constantly raised by partners; and (2) by the investigators’ own lived experiences as members of communities often “studied” but excluded from the planning and action of research: firearm owners and individuals from communities of color disproportionately burdened by violence. The investigators’ experiences of initiating the project illustrate the two positions from which researchers can enter into PAR with firearm owners: as an insider or an outsider (Ozanne & Saatcioglu, 2008).

Investigators as insiders are themselves members of those communities. PS entered into the project as a firearm owner. He is a white man who grew up in a rural area and in a firearm-owning household. He was socialized to think about firearms as a tool and part of recreational activities (i.e., target shooting and hunting). He was also exposed to cultural values that centralized firearms as distinctly American and pivotal to ensuring fundamental freedoms (e.g., The second amendment makes all the other possible). Additionally, he was informally taught firearm safety as a strong value and critical for responsible ownership and use. PS remains active in multiple communities that revolve around firearms, including hunting and competitive shooting events (e.g., International Defensive Pistol Association). He has also completed courses on the defensive use of firearms and has held an active permit to carry a concealed firearm. Representation of firearm owners on the investigator team is congruent with sharing power and elevating the perspective of firearm owners. Having a member of the investigator team who is a firearm owner and an active participant in the local communities involved in shooting sports had many benefits. These benefits included reducing distrust due to being a member of the same community, enabling recruitment of advisory board members and partners through personal relationships, and having a shared language and baseline knowledge. Community partners for the project reported that their willingness to partner with us was greatly influenced by knowing that one of the investigators was a firearm owner.

Outsiders, in contrast, are not members of the communities with whom they are partnering. Although KM entered into the project having grown up in an area affected by firearm-related violence, she was an outsider to the firearm-owning community. Having grown up in a high-violence neighborhood, KM had strongly aversive feelings about firearms. She was also ignorant about firearms: when KM started the work, she had no concept of calibers, different kinds of safeties, plinking, or variability in magazine capacities. Our community partners told her that it is important to make efforts to become knowledgeable about firearms, because ignorance could raise questions about her motives and commitment. Fortunately, the participating firearm owners were very willing to educate and socialize all members of the team (e.g., KM, student research assistants) about firearms. KM asked community members what she needed to learn, do, or experience to be a good partner. Based on their feedback, KM completed the NRA Basics of Pistol Shooting training course and joined firearm owners she knew at shooting ranges or in other firearm-centered activities. Those efforts to become knowledgeable created legitimacy and built trust; for example, our advisory board members commented that knowing that KM went shooting decreased their concerns that the project had hidden motives, such as that it would be used to further a liberal agenda.

Each investigator had existing relationships and community-based work; the urgent need to address firearm injuries was identified through KM’s partnerships with communities experiencing high violence exposure, but not through PS’s relationships with firearm owners. After CDC funding was released, PS and KM agreed to pursue the funding by collaborating to explore cross-cutting risk and prevention for all intentional firearm injuries, both suicide and violence. KM brought the notice of funding opportunity to her partners, who were enthusiastic about the opportunity to collect local information that could help to inform action, and they worked with her and PS to refine the ideas for this project. However, many of the relationships with the firearm-owning community were established for the purpose of this project. This process of developing the purpose, aims, and methodology among the two investigators represents one of the critical tensions, in a way, reflected by the utilitarian and empowerment approaches of PAR.

Upon receiving funding for the project, the investigators worked to build a team that was part of the firearm-owning community, consulting heavily with KM’s community partners. Building the team included decisions about how, where, and whom to recruit to join as key members of the team. As all decisions and activities aimed to flow out of the collaborative process among the team, these decisions were vital. Additionally, the investigators needed to decide how the team would be organized, define the roles, and generate a strategy for how power would be shared. The investigators identified two roles that would, in combination with the representation of firearm owners on the investigator team, reflect the partnership of firearm owners. These positions were Community Research Assistants (CRA) and a Firearm Owner Community Advisory Board (CAB). All CRAs and members of the CAB were members of the communities of focus.¹ The investigators team maintained responsibility to ensure that the work was conducted in accordance with ethical standards, Federal policy, the limitations of the budget, and best practices in quality data collection and documentation. Outside of those caveats, decision-making was shared across the whole research team, with CRAs in particularly fundamentally shaping the work. For an example related to the budget, in response to the CRAs’ input, the investigators doubled the incentives for interview participation, fought with human resources to raise the CRAs’ hourly rates, and paid to set up recruitment tables at sporting events. In best-practice PAR, ideally the community has full decision-making power. Researchers are responsible to share their knowledge about the existing state of research on the topic of interest, human research ethics, best-practice research methods, et cetera, but only have the final decision-making power for their own behavior. Our situation illustrates that external funds may exacerbate the utilitarian-empowerment tension.

The role of CRA was a half-time, paid position that was designed to ensure that the voices of firearm owners were integrated into the day-to-day functioning of the project. CRAs were firearm owners who attended weekly team meetings and were empowered to direct discussion, question investigators’ decisions, and lead components of the project. CRAs were also responsible for recruiting participants, collecting interview data, and engaging in analysis, interpretation, and dissemination. The Firearm Owner CAB was designed to be more consultative. CAB meetings took place monthly in the six months of the project, and then about quarterly after that. CAB meetings addressed specific issues relevant to the project at the time. For example, early CAB meetings were dedicated to identifying shared goals and developing the research questions and the methods for answering those questions, whereas later meetings were dedicated to ensuring representative participation, interpreting the qualitative data, and designing survey measures based on the qualitative data.

The investigators considered several logistic factors when partnering with our firearm owners around location and meeting schedules. The investigators aimed to identify spaces that were comfortable and convenient for our community partners. One the one hand, CRAs were formally hired and provided office space on campus with all meetings including CRAs and investigators taking place on campus. However, holding CAB meetings on the university campus would alienate people who may have concerns about universities’ agendas and ethics. Rather, the investigators partnered with the owner of a local shooting range, a former police officer who was recommended by one of KM’s community partners. The investigators went to his range to shoot and build a relationship with him as they were developing the grant proposal, and he agreed to be a partner; after the grant was funded, he opened his range to allow the CAB to meet there. Incidentally, the investigators gained two more members as people who came to shoot decided to stay and join the conversation, and then became permanent members of the board. CAB members told us later that the shooting range was much preferred over going to a university campus, as it reflected of our values of partnering with community members. The investigators also sought to identify a day, time, and meeting schedule that was acceptable to partners. The majority of our CAB members worked 9–5pm, and they preferred a 6:30 meeting time during the week. The investigators also provided meals and ate together to help build a sense of community, promote group cohesion, and develop social networks (e.g., Delormier et al., 2009); as an additional benefit, sharing meals predicts perceived well-being among older adults (Middleton et al., 2022).

Once the structure of the team and the roles were decided, the investigators next turned to the critical task of recruiting members into the team. The investigators aimed to construct the team intentionally and strategically to promote representation of the firearm owning community. Although one stereotypical image of a firearm owner is that of a conservative White man, firearm owners do not have a monolithic culture and are increasingly represented among various racial, ethnic, gender, social, and economic backgrounds. Having diversity while also sharing power in a manner that encourages conversation and promotes understanding different points of view is a vital part of PAR. The investigators attempted to honor and incorporate the range of perspectives, inclusive of White men but also of women and Black men, to promote exchange between individuals with varying experience and expertise, introduce novel perspectives, and encourage collaboration and respect for individual contributions (Chandanabhumma et al., 2023). A foundational value of PAR is to move beyond any single dominating group’s voice and to incorporate the voices of underrepresented groups in understanding the problem and acting. Failure to amplify the voices of people from marginalized groups hinders the exposure to new ideas and insights and limits what is possible. When recruiting members to join the team, the investigators constantly asked themselves and the community partners, “Who is missing from this table?”

In recruiting members of the team, the investigators hoped to recruit individuals with an eye to representation of diverse points of view (e.g., hunting culture, military/law enforcement, Second Amendment advocates, predominantly leisure/hobby use, predominantly protection use) as well as diverse identities (representation across race and gender). In addition to the goal of being able to construct social realities from diverse points of view, the investigators also wanted team members who could, and would be willing to, provide access to their communities for recruitment. Individuals who of their own social capital are well-networked, or who have influence in well-networked institutions, can, if they choose, provide points of entry into a community. When the number of points of entries increase to a critical mass, we believe that a community, in essence, has chosen to let the investigators in, which was one of our hopes for the project.

To do this, the investigators used multiple recruitment strategies; spending a lot of time together brainstorming messaging that would be accurate and appealing to individuals who had a sense of identity in firearm ownership. The final message had three main ideas: the investigators were looking for people to whom gun ownership was important; who thought that firearm-related injuries and deaths could be prevented; and who wanted to help figure out ways that firearm-related injuries could be prevented without violating the Second Amendment. Broad recruitment included formal and institutional means (e.g., television segments on local news describing the project; announcements through our university community communication system for both the CRA and CAB team members; job ads posting through general sites such as LinkedIn for the CRA positions). The investigators also used focused approaches, which tended to be informal and leveraged existing relationships with individuals or organizations; for example, talking to police partners; talking to and posting ads for both CRAs and CAB members at shooting ranges and community centers; using personal networks and friendships).

For CAB membership, there were about 15–20 regularly attending members; they included several people who learned about it on local news; several people who started attending after they saw the first meeting at the shooting range; two people who were friends of the investigators; two people who were connected to the police or military; several people who were connected to one of the partnering shooting ranges; and several people who were connected to the university. The group was racially diverse, including Black and White men, but White women were significantly underrepresented, and no Black women attended. Across the five firearm-owner CRA team members, two were connected to the military and four grew up in hunting culture; notably, two of our five community violence CRA members were also firearm owners, although they had not originally been recruited for that role. The CRAs were representative across race and gender. The majority of our CRAs came from existing partnerships and relationships; the investigators existing community partners knew how and where to find potential team members and offered introductions with an endorsement that conveyed legitimacy.

the investigators also aimed to onboard CRAs in a “cohort” of individuals. Though this was not always feasible, when a cohort of team members were brought in it made it possible to to build community through a shared experience of joining and becoming socialized into the team. The investigators also sought to explore each person’s motivations and goals for being part of the project and to express a commitment to help them move towards their goals—that is, to show that they see and respect potential team members as complex human beings rather than simply people to help get the work done.

When working with firearm owners, it is important to appreciate the strong likelihood that potential partners will greatly value privacy and confidentiality and may have particular concerns about how information will be used. Members of the team reported that they perceived the funding agency (i.e., the CDC) to be problematic and not trustworthy. Members of the team expressed concerns that Federal agencies have a history of promoting policies that reduce individual liberty, and they were concerned that Federal funding agencies such as the CDC might use the information generated in the project in a way that could threaten Second Amendment rights. Investigators should be transparent about their motives, the sources of their funding, and how the data will be used and protected. Interestingly, the investigators had a federal certificate of confidentiality, which offers a high level of assurance that participants’ and partners’ information could not be disclosed to others; yet community partners reported that this certificate was not important in their decision to join the team. Rather, they discussed that it was the transparency and respect showed by the investigators and that there was representation of firearm owners on the investigator team that made them feel comfortable participating.

Sharing power and co-constructing a team vision requires awareness of team members’ perspectives and development of intersubjectivity (i.e., sharing experiences, knowledge, and meanings from multiple perspectives in pursuit of mutual understanding). Such awareness raising and perspective sharing can be accomplished through several strategies. The investigators determined at the outset of the project that they would be open about their motivations and perspectives about firearms, firearm-related injury prevention, and the goals of the project, including our positionality and how that informed the project. This transparency typically involved the sharing our personal stories, background, and perspective with potential partners. The investigators regularly shared their experiences and relationship with firearms when interviewing CRAs, during CAB meetings, when doing interviews with journalists, and when presenting the project in professional and academic contexts. Owning their unique perspectives became a large part of the project. Such discussions aimed to foster transparency, investment, and partnership and a shared mission among the community partners.

Sharing power and recognizing the importance of each team member’s knowledge and experience is central to PAR. Although it was important that training was provided to community partners, the investigators approached orienting towards and developing the shared mission as one of cross-training: making explicit that knowledge and skill development is occurring in multiple directions. For example, it was necessary for the investigators to train CRAs about the research literature on suicide and violence prevention, human subjects research (including required certifications), PAR principles, and interviewing skills. It was equally important that CRAs educate the research team about their knowledge, skills, previous experiences, and their unique perspectives as members of the firearm owning community. We used several co-learning strategies that included structured experiences to direct sharing. Think-pair-share, for example, asked members to think on their own in response to a prompt, then turn and discuss with a partner. Pairs are then invited to share with the larger group. Given that the project included a qualitative interview component, having team members interview each other about their views and experiences with the aim of listening to understand promoted both skills development and increased understanding.

Investigators should also consider other educational and perspective-building strategies, including focus groups, conversations with key informants, and place-based engagement, among others. One strategy that was described by our team members as particularly valuable was the development and sharing of culturagrams (Congress, 1994). This approach, which arises out of family therapy, is a perspective-building strategy in which members of the team create a visual map of the culture of a group, including time in community; language spoken; health beliefs; trauma and crisis events; oppression and discrimination; religion and faith; holidays; food; clothing; value systems; history and contact with institutions such as legal, educational, and medical; and family structures. In the current project, members included additional topics covered in the culturagram should be something like relationships with firearms; conceptions of violence; conceptions of suicide. This moved towards multiple goals: we built team cohesion based on our willingness to be vulnerable with each other; our empathy for each other’s perspectives and values grew significantly; and we were able to have a window into worlds with which we had previously had no contact.

The relationships among the members of the team, both individually and collectively, were key to the success of the project. Many of the aforementioned strategies served, in part, the function of building positive relationships founded in mutual respect and trust. The investigators considered and discussed regularly how the conduct of all project activities could be carried out in a manner that also served to promote positive relationships and each team members personal and professional goals. We also used strategies uniquely aimed at building and maintaining positive relationships. Such strategies included sharing meals during and outside of meetings, activities related to firearms (e.g., going to the shooting range together), team social events, and broadly pursuing friendships and personal relationships. Admittedly, some members of the team were limited in their ability to devote additional to such events and activities due to life circumstances (e.g., being married with children, needing to hold multiple jobs). Such activities and events also often required financial resources, which required decisions about the best way to allocate available resources.

Investigators and team members were mindful to follow up on invitations issued by community partners as much as possible, including invitations to community members’ homes to see their firearm collections, attending and/or presenting at community events, and supporting partners’ organizations. The investigators also conducted formal quarterly individual check-in meetings that offered community team members time and space to discuss their experience on the project, offer feedback about the investigators’ leadership, and discuss their needs and how the investigators might help to address those needs. The continued solicitation of feedback from team members to the investigator team was essential to maintaining shared power.

The investigators also found that they needed to consider how team members could be supported given that they would be discussing and learning about highly difficult topics. Such support was part of the ethical considerations of the project and ensuring that there were distress protocols in place. Multiple instances of violence occurred throughout our project, both nationally and locally. Giving space for team members to discuss their reactions to these events, provide support to each other, and consider whether and how the work we were doing would be useful in reducing such incidents was essential to ensuring that team members were mutually supported and that our relationships positive. Relatedly, the interviews that CRAs conducted often involved learning about traumatic stories of violence and suicide. The investigators used grant funds to cover costs of psychotherapy at the psychology department training clinic for any team member who was interested due to concerns of vicarious trauma exposure and with the consent of the funding agency.

There were a variety of goals and associated tasks for the project due to the mixed-methods, multi-phase nature of the project. The necessity of such tasks and the timeline for completion, with one overarching objective being the successful conduct of the research, again reflects the tensions between utilitarian and empowerment approaches. Despite this potential discrepancy with the empowerment approach to PAR, the investigators attempted to perform these tasks in a manner that was consistent with PAR. For example, it was important to use generative strategies for brainstorming and throughout the stages and objectives of the project. Such strategies were used to develop ideas, solicit feedback, and review plans, practices, and procedures from multiple perspectives. Generative strategies included solo brainstorming and sharing with all individuals generating as many ideas as possible in a timed period. Ideas were then compiled into a group-level list and discussed. Group-level analysis asks members anonymously respond to prompts and then work in small or large groups to examine the resulting ideas, generate themes across them, and provide comments in an effort towards collective understanding and consensus (cf., Vaughn & Lohmueller, 2014). We used a particular form of these strategies often: the gallery walk, which developed out of educational research (Bowman, 2005). In gallery walks, individuals wrote their ideas on sticky notes and then placed them on the walls (i.e., on jumbo sticky post-it notes). After that, small groups walked around the room reviewing the responses and making notes on the items (e.g., making check marks for agreement, starring to emphasize, discussing themes across items) and then came together to discuss reactions and reflections.

One of the important early tasks for the project was to collaboratively develop a theory or logic model that guided the research questions, the specific methodology, and the data analysis plan. Our theory or logic model was informed by scientific knowledge and by the collective wisdom and lived experiences of our team members (Mehari et al., 2023). The research team and community partners worked to define the problem and goals and created a shared understanding of how that problem is developed, maintained, and how the processes of participatory action could create movement towards the goal. Though the investigators had largely determined the purpose of the project prior to bringing in additional members of the firearm-owning community, many of the specific details and decisions about how the project was carried out were constructed over time through conversations and generative activities and the initial logic model that continually revised as new information was received. This theory or model also informed the selection of the specific methods. For example, the advisory board members strongly indicated that interviews would be more appropriate than focus groups, due to the sensitivity of the topic and to reduce the likelihood that firearm owners would be unwilling to share their personal experiences and practices in a group setting.

The team also created new survey measures to assess constructs that emerged through our advisory board meetings and qualitative interviews in collaboration. To do this, we used generative strategies to create an item pool that reflected both the qualitative data and our team’s input. After generating items collaboratively, we used additional strategies to revise and refine the item pool. Sorting tasks, for example, involved printing out the items generated onto individual slips of paper and asking members to organize the slips by domains. Partners offered their view of how items did and did not go together. This was useful for identifying domains that may have been under-sampled (in which case, more items were developed) or oversampled (in which case, items were cut). This approach also revealed differential meanings of items across team members, which was useful for refining language. We also employed ranking and summing, which is useful for reducing item pools. This strategy involved team members ranking the importance of each item. Rank scores were then averaged, with the highest scoring items included in the final instrument. Team members were allowed to “rescue” items for discussion and propose revisions to items if they felt eliminating items would result in loss of important information. After item pools were generated and refined, we used a more participatory version of cognitive testing to continue to revise items. In cognitive testing, partners respond to the items as if they were a respondent. For our project, partners then came back together to discuss how they felt about the items more broadly, identify gaps (areas in the domain that were not assessed), and discuss concerns about bias or wording. A team member facilitated a conversation about the wording and meaning of the items one-by-one at the group level. Taken together, this process informed the revision of our items, scale organization and structure, and response options.

Once we collected our quantitative data, we co-analyzed data with partners to continue sharing knowledge and power, and to more usefully inform interpretation of the data. Co-analyzing data will look different based on the type of data to be analyzed. We used a grounded theory development and thematic analysis of qualitative interview data; employing sorting tasks and gallery walks, among others. We also shared our quantitative data, following de-identification, and emerging results presented continuously as part of the participatory process, often through data visualization such as bar graphs and pie charts. Such co-analysis of data aimed towards member checking, aims to improve the accuracy and credibility of research findings. Investigators shared emerging themes of qualitative data and the results of quantitative statistical testing and ask community partners for their critical feedback. Community members checked biases, suggested alternative thematic structures or analytic approaches, clarified meanings, and helped interpret the data.

Once data were collected, analyzed, and interpreted, the tasks turned to disseminating findings. The researchers and community partners collaboratively developed a dissemination plan that included scientific, public, and local dissemination, striving to meet the needs and values of the researchers and community partners. This dissemination looked different across types of partnerships. For example, part of the data we collected were from local school systems. Accordingly, we were sensitive to the reality that our school partners would have major concerns about disseminating information regarding firearm-related violence among their students. Data were presented to school administrators and it was discussed how we would use data collected in their school. We provided school administrators with the only copies of school-identifiable data summaries. Investigators should not disseminate information unless the community partners agree, and, ideally, at least one representative review all the dissemination materials beforehand.

There also needs to be have transparent conversations about what the investigators need as part of their jobs (e.g., scientific publication for tenure or promotion) and what the community partners need or want related to dissemination. Community partners may want to take the lead on media dissemination (e.g., news, podcasts), may want to serve as authors or presenters on scientific dissemination (manuscripts, chapters, conference presentations), or may be interested in ensuring that the data is shared locally with invested or relevant individuals, agencies, or institutions. Community partners can provide information about how the data could be used locally and with whom and how the data should be shared. Non-identifiable, aggregated data could be shared with institutional partners to answer their questions or provide information about the strengths, needs, and functioning of their institution or constituents/consumers/clients.

Investigators and other team members may possess biases that make a PAR approach challenging—many people want there to be a right and wrong, but a fundamental principle of PAR is that there are many truths, not just one. “Check your bias” is a common refrain for engaging in both participatory and qualitative research, but how to actually do that always seems to sink into vagueness, covered with words like “reflexivity.” We attempted to check our biases by creating a team full of people who had different perspectives and were willing to disagree with each other, including and perhaps especially the investigators, in a group setting. The investigators were mindful both about having racial, ethnic, and gender diversity on the team as well as diversity of opinion. It was critical to our success that the team members were able to accept and hold the tension that resulted from that diversity. Indeed, a lack of tension or disagreement on a team is a sign that it may have never been adequately diverse to start out with, or that it moved into groupthink (e.g., Kidd & Kral, 2005).

Disagreement is useful but difficult to navigate, especially when the invested parties have a strong need for their opinion to be agreed upon as the “correct” one. Disagreement can include ideologically opposed issues such as violence as a useful tool or means to an end versus an evil to be avoided; justifications of police use of deadly force; the extent to which homicides can be justified, if it all; attributions of violence and dangerousness to specific racial groups; stigmatization of people with mental health difficulties; rationales behind violence against women; and many other sensitive, highly opposed ideals. During intense disagreements among team members, the investigators reminded the group immediate goals of the meeting as well as the broader mission of the work. Specifically, the investigators reiterated that part of the work is to understand different perspectives deeply; the purpose of this messaging was to shift partners from trying to persuade each other to examining how we could craft questions or survey items to understand differences of opinion, regardless of the accuracy of those opinions. Emphasizing the shared goals of reducing injury and death, across political ideology, also helped to refocus conversations. Other de-escalation strategies that we periodically needed to employ included investigators talking one-on-one with CAB members to understand their perspective and express appreciation for their participation and breaking into groups strategically (i.e., separating people who tended to be most likely to enter into strong disagreements with each other or pairing them with advisory board members with whom they had good relationships). Such approaches were largely successful in maintaining or returning to civil discourse. We do not advocate for expelling members or cancelling meetings. Embarking on PAR with firearm owners is to accept that difficult conversations will be had, and the whole team must be committed to having them.

In addition to the disagreements within the CAB, the investigators and CRAs were also at times exposed to strong and fundamentally opposed beliefs that caused serious distress; for example, during interviews or in community meetings. The team found it useful to engage in two strategies: developing safe spaces where we could share our visceral reactions to ideas that could cause harm to some of our communities, and focusing on shared goals. The investigators decided that in the CAB meetings, they would be authentic but not fully transparent. One reason was so that the investigators would not be taking up disproportionate space; another was that our deepest emotional truths should be reserved for our most intimate relationships. So, for example, KM disclosed that she grew up with an aversion to and fear of guns due to neighborhood violence, but she was not transparent about her fury and grief around police-involved shootings. However, in the research team meetings that included the investigators and CRAs, the team agreed to be transparent with each other about their emotional experiences, which allowed us to provide support and processing for each other when we encountered ideas that were difficult to hear. In addition, as a team and partnership, we consistently reminded ourselves what our goals are: to prevent injuries and save lives. Every single person on the project was highly motivated to move towards those goals, even though we sometimes disagreed about the best way to do so. Reminding ourselves that everyone wanted to make their communities safer made it possible to let go of some of the immediate and strong distress reactions.