Diversity, Equity, and Inclusion from the Ground Up: Evaluating the Impact of a Community-Based DEI Intervention

This study presents an evaluation of a film screening and interactive panel created and presented in cooperation with multiple community stakeholders. This program, which we are labeling a diversity, equity, and inclusion (DEI) intervention, was designed to open channels of discourse with and about the local Somali population in a rural, predominantly White midwestern region to disrupt the pervasive and negative depictions of Somali Muslim immigrants that dominate mainstream media. Our goal for evaluating the intervention was to better understand its effectiveness in increasing understanding and empathy for the experiences of Somali immigrants and for increasing the potential for positive interactions between county social service employees and their Somali clients and neighbors. Our analysis showed the intervention effectively raised the largely White audience’s consciousness regarding their own perspectives and biases. This led to increased perspective-taking and feelings of connection, which can be key antecedents to increasing positive interactions.

As populations around the world continue to diversify, communities face pressures to overcome conflict and capture the potential benefits of inclusivity.Higher-education practitioners and researchers face new opportunities and obligations to help their communities negotiate these changes by improving tools to advance equity and inclusion.In this paper, we report on a collaborative project that connected residents of a rural Midwestern community with government, non-profit, faith, and higher-education stakeholders to develop and deploy a locally sourced diversity, equity, and inclusion (DEI) intervention.The intervention paired the screening of a purposemade documentary film with an interactive panel to connect members of the predominantly White community (about 90% according to the U.S. Census Bureau, 2022) to perspectives from firstand second-generation immigrants from Somalia, a prominent and growing minority population who often practice the Muslim faith.Our analysis indicates that community-based research and action collaborations carry the unique potential to enhance the desired outcomes of DEI interventions.

Diversity, Equity, and Inclusion Training
DEI training broadly describes a variety of organization-based instructional efforts, both experiential and information oriented, meant to minimize conflict in heterogeneous organizations by facilitating positive intergroup interaction, reducing prejudice and discrimination, and enhancing skills and motivation to interact with diverse others (Bezrukova et al., 2016).Many DEI initiatives find theoretical grounding in the intergroup contact theory, sometimes called the social contact theory, which suggests that prejudice between members of different groups can be reduced when they come together under optimal conditions of equal status, shared goals, personal intimacy, and sanction from authority (Paluck, 2006).Research on intergroup bias indicates that positive interpersonal contact across ingroup and outgroup boundaries can reduce stereotypes and bias (Hewstone et al., 2002), in part by encouraging cognitive dissonance between those stereotypes and new information intergroup contact can bring.The drive to resolve that dissonance can lead to the productive modification of stereotypical mental categories (Leippe & Eisenstadt, 1994).
One common DEI training approach aims to increase empathy via perspective-taking, which breaks down barriers between in-and outgroups (Lindsey et al., 2015).Actively discrediting stereotypes also plays an explicit role, though strongly held beliefs linked to intergroup divisions (e.g., race and gender) may cause resistance to this method (Lindsey et al., 2015).With the rising prevalence of TV and film in the 1950s, scholars in psychology observed that the emotional bonds audiences form with fictional characters and celebrities could evoke effects similar to real-world relationships.This phenomenon was dubbed "para-social contact" (Horton & Wohl, 1956), and media-oriented research on the parasocial contact hypothesis has expanded in the decades since (Bond, 2020;Park, 2012;Schiappa et al., 2005).In the context of DEI training, researchers in one study of 239 participants found that viewing a 30-minute documentary evoked effects on empathy and prejudice reduction similar to those after a live 4-hour training (Cadenas et al., 2018).Taken as a whole, the research suggests that DEI interventions utilizing both social and parasocial contact dimensions hold promise for reducing prejudice, disrupting stereotypes, and enhancing empathy toward outgroups.
Researchers have also identified shortcomings of DEI training, including limited effects on behavioral change, the risk of masking inequities, and increasing privileged groups' perceptions of victimhood (Chang et al., 2019;Dover et al., 2020).Some recent analyses suggest that enthusiasm and investment for DEI training have outpaced evidence of its effectiveness (Devine & Ash, 2022), especially with mandatory trainings.The professionalization of DEI consulting has also raised suspicion of a "DEI-industrial complex" (Read, 2021) taking hold in which organizations and professional trainers create a mutually beneficial and self-serving feedback loop that gives the appearance of inclusion efforts but has no real impact.
Still, the demand for DEI training remains high today for the same reason it has for decades: the pressing need to respond to society's increasing diversity (Ehrke et al., 2014).One communitybased model of DEI intervention that may be instructive is the Safe Zone/Safe Space approach to institutionalizing support for LGBTQ+ populations.Originating in the 1990s, Safe Zone programs are decentralized ally training initiatives designed to establish grassroots networks of LGBTQ+ support particularly among nonqueer members of a community, most often on college campuses (Draughn et al., 2002).Unlike professional DEI consultants or one-size-fitsall stock training modules, Safe Zone and other ally-oriented programming relies on several key elements to foster a representative and inclusive intervention.These include direct involvement of the relevant underrepresented community in design and delivery of training (Lorenzetti, 2010), open dialogue through affirming conversation techniques (Poynter & Tubbs, 2008), and framing the opportunity as personal development rather than crisis intervention (Gardner & Alanis, 2020).This community-situated, ally-building approach can help overcome some of the common problems with typical DEI trainings, such as the defensiveness majority populations feel when confronted with evidence of bigotry and a suspicion of experts "parachuting in" to fix things.Research suggests that allyship as an inclusion tool (in contrast to didactic "top-down" diversity training) can yield a range of benefits, particularly in organizational settings (Salter & Migliaccio, 2019).When invited to be allies, especially by coworkers or fellow community members, trainees see themselves as part of the solution rather than part of the problem (Gardner & Alanis, 2020).With an emphasis on personal relationships and local context, there is a risk that ally-oriented DEI interventions lead participants to overlook the systemic origins of oppression (Broido, 2000).Despite this critique, we suggest that ally-oriented DEI interventions have the power to improve intergroup connections and increase support for difference among majority populations within a community.These are necessary, though not sufficient, elements in moving toward lasting change.
Beyond its alignment with key features of allyship training, the modality of the intervention under analysis also differs in significant ways from traits of some popular implementations of DEI training that have drawn criticism, such as the emphasis on theory and broad principles (e.g., anti-bias training) and the use of outside "experts" to deliver content to passive audiences.The "film and conversation" format of the intervention under analysis is interactive rather than didactic, rooted in community knowledge rather than general concepts, and guided by the perspectives of local people, particularly those from the underrepresented population in question, rather than outsiders.To acknowledge both the similarities and differences with standard models increasingly shaped by the so-called DEI industrial complex, we refer to the object of study as a DEI intervention rather than a DEI training.

Theoretical Framework
This project was guided by a communitybased participatory action research (CBPAR) framework.The central principles of CBPAR include partnerships with community members as collaborators engaged in democratic processes that aim to eliminate inequity and advance social justice (Zuber-Skerritt, 2018).CBPAR has been used by health researchers who recognize that community members bring strengths and make meaningful contributions as research partners (Mocarski et al., 2020).As will be described more fully below, the process of developing and deploying the DEI intervention followed CBPAR principles.Across the stages of the project, the researchers collaborated with community members with the aim of identifying and addressing problems relevant in this community context (Greenwood & Levin, 2007).Members of the community were integrated into the project as knowledge partners (Stern, 2019) to better comprehend the lack of understanding of Somali immigrant experiences and strategize pathways toward positive change.Representatives from across the community, largely members of stakeholder organizations such as support agencies and the Islamic center, contributed to identifying the project goals, establishing the lines of inquiry, providing and interpreting relevant perspectives, and communicating the results with the community at large.Additionally, the project aimed to build capacity for change by fostering social networks that can carry "ripple effects" throughout the community ecology (Trickett & Beehler, 2017).
Unlike a traditional research project that might yield a written analysis or formal presentation (or even a standalone documentary film) as its final output, community input steered this project toward an iterative and interactive format we that are calling a DEI intervention.When referencing this CBPAR project, we are referring to the whole of this work-from initial inquiry to its ongoing implementation.It has involved community partners of many backgrounds, including but not exclusive to the vital collaboration with members of the Somali community.
This study presents an evaluation of one key goal of our CBPAR project: the impact of the DEI intervention on the attitudes, values, and beliefs of the majority (in our case, White non-Muslim) population.Ultimately, the project's true measure is its impact on the lived experiences of Somali community members.Though some DEI training theory suggests that positive changes in majority population attitudes should yield parallel improvements in the experiences of minoritized populations, the survey of literature provided above shows that the research is inconclusive on this point.Focused and specific assessment is necessary to understand the outcomes for Somali residents.We have not examined those outcomes here because it requires evaluative techniques beyond the scope of the tools utilized in this analysis.However, because changes in behaviors or policies affecting minoritized populations are unlikely to occur without antecedent changes in attitudes and beliefs, it is valuable to understand how and why community-based interventions can contribute to those changes.In the next section, we describe the relationship development, project planning, and ongoing discussion that took place throughout the project.

Background
The collaborative DEI intervention described and analyzed in this paper was developed in a rural Midwestern community of about 50,000 people with a fast-growing number of families from east African nations, particularly Somalia.This project began as a response to public calls for action in this community amid a rising tide of bigotry and even violence against Muslims and immigrants at local, state, and national levels.In the first phase of the project, a local university professor (first author) engaged his established community networks to meet with local stakeholders and recruit collaborators to the project.Various perspectives on the issue were collected over several weeks, shaping the nature and scope of the overall project.
In the second phase, the first author and an undergraduate student co-producer collaborated with dozens of local Somali participants to produce a short documentary film.To contrast with negative narratives dominating the public consciousness, our Somali partners emphasized the need for a fuller range of stories to be told about Somali life and experiences, from the intense struggle of establishing a new home to their optimism for future generations.In recent years, depictions of immigrants in American popular culture have been consistently narrow and negative (Bleich et al., 2019;Omidvar & Richards, 2014).Negative frames pathologize immigrants, especially Muslims, depicting them as terrorists and religious extremists (Zakaria, 2016) or as an invasive threat to the nation (Cisneros, 2008).These representations rarely include success stories, self-expression by the group in question, or depictions of similarities between the immigrant and majority populations.By framing Muslim and immigrant populations in these ways, media representations can establish durable stereotypes with harmful effects (Borah, 2011;Entman, 1993;Volkmer, 2009).
As non-Somali people, the documentary coproducers (one identifies as White, the other as a mixed Anglo-Latino child of an immigrant parent) practiced critical reflexivity to help navigate the ethical issues of voice and perspective in depicting stories of a marginalized group (DeGroof, 2013;Ginsburg, 1995;Ruby, 1991).They followed the lead of their collaborators to ensure that Somali voices and experiences resonated through the film in ways their Somali peers would recognize as authentic.Ultimately, the producers sought what researcher and filmmaker Barbara Myerhoff called a "third voice, " where the voices of the subjects become embedded in that of the filmmakers, such that an audience cannot distinguish one from the other (Kaminsky, 1992;Sudbury, 2016).
In the third phase, the collaborators settled on the DEI intervention format as the most promising method to communicate the product of our research to the community at large.The intervention featured a screening of the documentary followed by a dialogue between a panel of Somali residents and the audience of (generally) non-Somali locals.Various community organizations, some of whom had participated as collaborators in the first and second phases, hosted these events.Between 2018 and 2022, this intervention has been presented 20 times to over 1,200 people.In recognition of the labor our Somali collaborators have provided to educate their non-Somali neighbors, those whose contributions can be reasonably calculated (e.g., by participating on the discussion panel) have been paid between $20 and $40 per hour for their work.Funding was provided through a small community-engaged research grant provided by the first author's home institution.
All phases of the project featured close collaboration with stakeholders from across the community including government offices, faith groups, nonprofit support agencies, community leaders, and invested individuals.In line with best practices outlined by Clark and Missal (2017), the process placed particular emphasis on strengthening and extending trusting relationships with members of the Somali community, whose involvement played a defining role in shaping the project.For example, the idea to pair the film screening with an interactive panel of local Somali residents came in response to expressed concerns from Somali collaborators that the documentary may not represent the experiences of some Somali people.This mechanism provides opportunities at every event for Somali panelists to contradict, reinforce, or expand upon the perspectives shared in the film.
Driven by community input, this DEI intervention was designed to expand the available narratives of immigrants and Muslims, particularly Somali Americans, to disrupt the toxic and narrow representations offered in mainstream media that may contribute to the alienation many local Somali residents experience.The screening and conversations bring forward a new set of stories to invite a fuller exploration of minority perspectives.They encourage an invitational exchange of ideas, offering a constructive opportunity for shared meaning-making.They also allow Somali-Americans to be creative agents of their own storytelling, a stark contrast to the reactive, defensive stance that is often forced upon minority groups-especially Muslims-when confronted with the pathologizing media frames (and the personal attacks they sometimes trigger) that have become the default in Western societies.

Methods for Assessment
With this study, we sought to establish the effectiveness of the DEI intervention described above as deployed for Human Services staff members of two local county governments.Using individual interviews, we investigated the impact of the intervention on the staff 's perceptions and attitudes.We also identified the design elements of the intervention the staff regarded as most impactful.With consideration for the principles and goals for effective DEI training, as well as the more specific goals of our community collaborators, we used the following questions to guide our evaluation: 1.In what ways did the intervention affect participants' perceptions and attitudes toward their community's diversifying population? 2. Which specific aspects of the intervention most affected participants' perceptions and attitudes?
In keeping with the participatory action framework used in the development of the intervention, we approached data collection with a focus on the needs of our community collaborators.As described further below, we used qualitative methods to evaluate the outcomes of the intervention, collecting data through individual interviews.To respect the desires of the Human Services agencies, we limited our participant recruitment, relying on supervisors in those offices to recruit staff willing to be interviewed.The length of the interviews was also limited by the requirement to complete the interviews during the workday.Although these circumstances are far from ideal in gathering a representative amount of data from which to extrapolate findings, we accepted this limitation as a condition of the working relationship with our community partner.

Participants and Procedure
Human Services staff from two rural counties participated in a screening-and-discussion event.Eighty-five staff were divided among three sessions to increase opportunities for interaction.The intervention began with the aforementioned 35-minute documentary presenting experiences and perspectives from members of the local Somali community.This was followed by a 90-minute interactive conversation with two local Somali volunteers (one of whom appeared in the film) who shared their own reactions to the film, their experiences living in the community, and their responses to audience questions.
We used qualitative methods to evaluate the impact of the intervention on participants' perceptions and attitudes.Several months after the intervention, we recruited individuals to participate in follow-up interviews.To respect privacy and accommodate organizational constraints, we recruited interview participants in the county agencies through their supervisors and conducted the interviews during business hours.This approach could have artificially limited the length of the interviews or constrained who felt comfortable participating.Through the supervisors, we recruited 11 interview participants, which we considered an adequate sample size.Using a semi-structured interview protocol (Appendix A), the second author conducted audio-recorded interviews in a private room at the participants' workplaces during the workday.Prior to the interviews, each participant provided informed consent pursuant to protocols evaluated and approved by the Gustavus Adolphus College Institutional Review Board (approval #1819-0105).
The interview began with a brief introduction, a statement of confidentiality and privacy, and an affirmation of consent to record.Participants were asked to respond to the following questions: needed to answer the questions; however, because the interviews took place during the workday, the interviewer asked minimal follow-up questions.The researchers opted for this compromise out of respect for the participants and the partner organizations.The 11 interviews were conducted six months (n = 6) or eight months (n = 5) postintervention.The length of the interviews ranged from 3 minutes and 45 seconds to 25 minutes and produced 104 total minutes of audio.Despite the limited amount of data, the participants' reflections provide valuable and unique insight into their lived experiences that would otherwise remain hidden (Stern, 2019).
The recordings were transcribed by the second author and a research assistant and anonymized before analysis began.All transcripts were analyzed by both authors.We used an inductive approach to carry out a rigorous analysis of the raw data (Thomas, 2006).We assigned descriptive, content-based codes to the data (Saldaña, 2016), then reviewed the transcripts and codes together, discussing unique and similar codes and condensing related codes.We moved from descriptive coding to analytic coding and mapped the relationships of the coded categories to one another (Saldaña, 2016).In the following section, we report on the themes that emerged from the analytic coding.Our analysis was guided by the goals of the evaluation (Thomas, 2006), specifically seeking insight into participants' perceptions and attitudes toward their community's diversifying population, how the film and panel affected their perceptions and attitudes, and which specific aspects of the film and panel had the greatest effect.

Results
Perhaps the most prominent theme throughout our analysis is the value of personal connections.Participants universally felt that learning the stories shared in the film and engaging in face-to-face dialogue with Somali participants from the community was meaningful.Participants valued the simple act of exchanging direct questions and answers, and several expressed the desire for more time to do so.For some, the interaction with Somali people advocating on their own behalf, whether in the film or in person, was a rare and valuable experience: "It was nice to have the actual people in the video and speaking versus [a non-Somali person] speaking on it, saying, 'We interviewed these people.' [It] was nice to have people explaining their own side of things" (Participant 4).
Beyond the personal connection to Somali people, participants noted that the film's reliance on local context contributed to its impact-not only the inclusion of local people but also familiar places and issues.Allusions to current events, clear knowledge of local customs and culture, and reference to common landmarks made the film more credible and immediate.Several participants remarked that the localized context increased their engagement: "This was super helpful to actually hear actual stories from folks who are here in the community.… To hear those stories, I think, actually affects people more" (Participant 10).This comment illustrates how parasocial contact can create bonds and yield effects similar to interpersonal effects (Bond, 2020;Schiappa et al., 2005).
Whether attributed to the local context or other factors, every single interview participant noted that the intervention provided an engaging experience for them.Though some degree of social desirability bias may be in effect, with participants feeling they "should" say positive things about the film and panel, the nature of the feedback suggests they were substantive and authentic expressions.Some described their desire to bring the film screening and panel to other organizations of which they are part, and one participant placed it at "the top of the list" (Participant 11) for diversity trainings they have experienced.Many stated they "enjoyed it" or described it as "effective" or "really good" (Participants 1, 5, 6, 8, 9).One participant stated: I was glad that I went.I felt after leaving, you feel a little heavy, but then it's a good heavy, it's a good, 'Okay, now what?' I think when we know better, we do better, and that's what I took from that.(Participant 6) Participants reflected frequently on ways the film and panel interaction challenged stereotypes about Somali people and culture held either by them personally or by society at large: "I was under the impression [Somali people] hated us.
[The intervention] kind of took away some of the stereotypes that I have for the most part, and [showed] that they really are like us, just wanting to get along" (Participant 2).Some participants spoke specifically to the cognitive dissonance they felt when the content contradicted their previously held beliefs: I guess, ever since then, that big thought that I brought away has kind of put my mind and thoughts and beliefs about everything in total turmoil.Now I think my whole perception has changed a little bit, and now I don't really know what to think, to be really honest.(Participant 5) As we will address in the discussion, this dissonance can play an important role in interrupting stereotyped thinking and moving a person toward seeing their inherent connection to outgroups.
Many participants described how the film and panel exposed common ground by providing new insights about Somali people and their lives.Variations on the refrain "They're not so different from us" were noted throughout the transcripts.Some participants also reported specific points of connection, including faith ("Our religions aren't all that different, " Participant 5) and motherhood: [The Somali woman] made a comment about being worried about her kids … and just the feeling of you would do anything for your kids.And that right there is what made the connection for me.Yep, we are very alike.I don't care what color you are, the feeling that you have that you would do whatever it is to protect your kids.That's the one statement of everything that made me realize we are more alike than we realize.(Participant 2) Participants described shifts not only in intellectual understanding but also toward deeper emotional connections to the experiences of their Somali neighbors.Expressions ranged from subtle observations to substantial epiphanies indicative of growing empathy, an important component in eliminating barriers between ingroups and outgroups (Lindsey et al., 2015): We were listening to a presentation in our full staff meeting the other day, and [we were asked] to think about things that have really impacted your life when it comes to looking at things differently and trying to understand it from other people's eyes.The intervention also moved many to think deeply about their own feelings, attitudes, reactions, and behaviors: [The intervention] just blew my mind and it really made me more aware of yeah, how do I feel?How do I take that in?How am I possibly judging?Do I try to seek to understand?Do I even have, personally, people from other cultures in my life?(Participant 6) I'm not sure how much of a better understanding I gained from it, but certainly, maybe less fear of inapproachability … just because of having more in common than I had really thought about before.(Participant 7) For some, these revelations had an explicit connection to their professional work as county employees, largely in social service roles: There's just going to be more potential friction because even in my job, there are cases where there might be misunderstandings because of cultural differences.And so I think just the more … chances we have to actually talk about it head-on, the better.I hope more trainings, more offerings to view the video would be available.(Participant 10) The growing presence of Somali people over the past 30 years has evoked many questions and reactions among residents of the community, some of them seemingly taboo.Several participants registered appreciation for a safe space to ask questions and get firsthand insights from people with direct knowledge on topics including Islamic customs, government assistance, modes of dress, and food preferences: I think on the surface people from different countries or different cultures, they seem so different or estranged and you don't understand as to why and you are kind of intimidated and you don't really know how to get those questions answered, but this was a really effective way of … you get to see that these people have the same cares and thoughts and ideas that you have.(Participant 9) To experience [the conversation] from a parent's perspective was very interesting to me, because I felt more comfortable talking to my kids when … somebody looks a little bit different, dresses a little differently, those kinds of things.(Participant 3) Two interviewees, including one who selfidentified as a member of an underrepresented group in the community, found special significance in enabling these conversations.Taking part in discussions about equity and inclusion issues, even simply witnessing others do so, provided hope for positive change in their workplace and community: "So I think for people like me who maybe experience some things sort of like that [racial discrimination], it's a little bit affirming to know that people are likely thinking about it and hearing about it" (Participant 10).
While the interviews documented many positive experiences, one participant offered a reaction that contrasted sharply with the others in a particular way.In addition to acknowledging new and helpful insights from the event (among them the realization noted above that Somali people do not "hate us"), this person reacted negatively to a perceived message that the majority culture needs to be welcoming and accepting without expecting reciprocity from Somali people.Specifically, this individual was seeking assurance that refugees would adapt to the majority culture and strive to integrate more actively into the community at large.Although these views were not expressed by other interview participants, it is possible, even likely, that others who participated in the intervention might share them.This exemplifies some of the possible shortcomings of DEI training, particularly the risk of increasing privileged groups' perceptions of victimhood (Chang et al., 2019;Dover et al., 2020).

Discussion
In this paper, we explored the potential to disrupt stereotypes and prejudices through a community-based, collaboratively developed DEI intervention that allowed Somali and non-Somali members of our community to forge new connections grounded in perspectives of the Somali population itself.The intervention was produced in partnership with local stakeholders using principles of participatory action research, which seeks to leverage the perspectives and insights of those whose situation is being investigated by actively involving them in the research process (Stern, 2019).Through an evaluation of the intervention, we learned that collaboration among local community stakeholders is valuable and effective in creating opportunities geared toward inquiry and transformation.
Our findings suggest that DEI trainings designed and implemented in connection with community stakeholders, especially local members of underrepresented groups, can facilitate the beneficial outcomes for which such interventions are designed.Integrating stakeholder perspectives into the design process infused this intervention with content and credibility that helped participants engage with difficult social problems such as racism and anti-immigrant bias.The community-based origins of this intervention seemed to provide a level of credibility that persuaded participants to acknowledge and cope with, rather than minimize or dismiss, the vital seed of dissonance necessary to challenge stereotypes and find empathy with their Somali neighbors.Through our data collection and analysis, we sought to answer two research questions.The first related to the intervention's impact upon participants, the second to qualities of the intervention itself.

Positive Interactions Disrupt Stereotypes and Encourage Empathy
For our first research question, we asked, "In what ways did the intervention affect participants' perceptions and attitudes toward their community's diversifying population?"Our data suggest three key areas of impact relevant for DEI training effectiveness.The first and most compelling is the disruption and discrediting of stereotypes.Disrupting stereotypes is well-established as a central pillar in effective DEI training (Lindsey et al., 2015), and the results of our study suggest participants experienced disruption through both direct and indirect means.Some participants described a direct effect whereby information presented plainly and persuasively contradicted the ideas they previously held.Others expressed a less direct but perhaps more pervasive disruption of ideas in the form of cognitive dissonance.Research indicates that experiences such as these play a key role in unseating the deeply held mental categories that can make stereotypes so resistant to change (Leippe & Eisenstadt, 1994).
Second, participants in this intervention experienced a high degree of positive intergroup interaction, a method significantly correlated to reducing prejudice and encouraging positive future interactions (Bezrukova et al., 2016;Hewstone et al., 2002).Positive interaction was documented in multiple ways, including (a) expressions of gratitude for the film and panel and for the Somali experts in particular, (b) remarks that the panel simply was not long enough to satisfy the urge for interaction with the Somali residents, and (c) comments conveying the desire for others to experience the film and panel.Third, participants articulated feelings of empathy and perspectivetaking.For example, some participants expressed an increased ability to relate to the unique experiences of their Somali neighbors.Other participants forged connections by discovering common values and experiences.
Though not sufficient in and of itself to effect the systemic changes necessary to overturn social dynamics of oppressive power, establishing shared humanity and empathy for the experiences of people with diverse backgrounds plays a major role in reducing barriers between ingroup and outgroup populations (Lindsey et al., 2015).Moreover, in identifying how and why their thinking changed, many participants demonstrated a capacity for reflexivity about race from which Whites have been historically shielded.We find hope in observing that the intervention invited its mainly White audience to take note of the malleabilityand even the mere existence-of their own attitudes toward a minority racial-cultural group in their community.This critical self-awareness is both a major component in the process of the intervention as well as a key outcome, as described in the next section.

Local Contact Fuels Self-Awareness
For our second research question, we asked, "Which specific aspects of the intervention most affected participants' perceptions and attitudes?"While our analysis indicated that many elements of the intervention contributed to its overall effect, two themes stood out strongly: forging personal connections and critical self-awareness.Based on data from the interviews, forging personal connections with Somali neighbors-both through direct social contact (via the face-to-face interactive panel) and parasocial contact (via stories presented in the documentary film)-contributed to attitude changes, empathy, and other beneficial effects.Multiple respondents emphasized the value of hearing testimony provided directly from the Somali community.Even several weeks after the intervention, nearly all interview participants displayed vivid recall of anecdotes related to the film or discussion panel.Some participants indicated that the interpersonal interaction was the most affecting element and lamented not having more time for it.These clues suggest that these new personal connections carried high potency.
A small number of participants noted a limitation stemming in part from the personal interaction format.They observed a hesitance (their own or that of colleagues) to ask questions freely in the semi-public forum for fear of offending the panelists or, more pointedly, being perceived as ignorant or insensitive by their peers or supervisors.This feature may also inhibit participants' abilities to openly discuss negative experiences or attitudes regarding their Somali neighbors, preempting opportunities to further defuse stereotypes and misconceptions.Such constraints could limit the intervention's efficacy in reducing prejudice and should be addressed in future training environments, perhaps by offering an anonymous method for submitting questions or comments.
Overall, our findings regarding personal connections align with DEI training theories based on the social and parasocial contact hypotheses, which suggest that positive contact across ingroup/outgroup boundaries can yield important benefits (Cadenas et al., 2018;Paluck, 2006;Schiappa et al., 2005).Our results also align with research indicating that vicarious contact (as with a documentary or workshop) can enhance feelings of empathy and reduce prejudice toward outgroup members (Cadenas et al., 2018).During their interviews, multiple participants reflected on the impact of the personal stories shared both in the film and during panel dialogues, citing specific examples from each.Though the methodology of this study does not allow us to parse the relative impact of the live conversations versus the documentary film, our results indicate that both played a role in moving participants to recognize the common ground they hold with their Somali neighbors.The local nature of the connections also contributed to this impact by offering higher familiarity and credibility than generic sources such as national news outlets or mass-marketed training modules.Research suggests that establishing the local credentials of the outgroup population may reduce stereotyping by diversifying the majority population's idea of who composes the perceived ingroup of "local people" (Ehrke et al., 2014).
A second theme in the data points to the intervention's effectiveness in stoking critical self-awareness among participants.For many participants, this reflexivity was initially triggered by the stories and opinions shared by the Somali people in the film and panels.Often, these insights in some way contradicted the ideas about immigrants and/or Muslims that participants previously held, pushing them to grapple with the conflicting narratives held in their minds.By introducing cognitive dissonance, interventions like this can provide a catalyst for constructively modifying the biased mental frameworks on which stereotypes are built (Leippe & Eisenstadt, 1994).
Consistent negative portrayals of Muslims and immigrants in mainstream media present a powerful and pervasive image that frames these groups as a problem in society (Bleich et al., 2019;Omidvar & Richards, 2014;Zakaria, 2016).The stories shared in the film and conversation contradicted these pathologizing narratives, introducing a dissonance that heightened participants' sensitivity to their misperceptions of the outgroup.This reflexive thinking moved some to scrutinize extant biases (their own or society's at large), while others overtly questioned their role in perpetuating narrow, misleading narratives about Somali people.This process of critical selfawareness can contribute to positive attitude change and disruption of the negative stereotypes about immigrants.

Reactions from Somali Partners
The lived experiences of the Somali individuals and groups who served as collaborators in this project shaped its goals and trajectory in indelible ways.While this study officially examines the impact of the community-based DEI intervention on its intended audience (mainly White non-Muslims), it is also important to recognize the Somali perspective, particularly in revealing some of the effects outside the scope of this analysis.
As university faculty, the co-authors of this study possessed resources to conduct a formal assessment that non-academics generally do not, so our community collaborators were less able to participate directly in the assessment process.However, to glean some insight into the effects among our Somali neighbors, the co-authors engaged a recognized leader in the local Somali community who played a key role in developing and deploying the intervention.In a joint conference presentation where the co-authors offered a preliminary version of this paper, this local leader contributed his reflections on Somali community reactions and outcomes.Though difficult to assess whether interactions with county employees had materially improved, he expressed confidence that the intervention promoted understanding of the Somali culture and that the direct involvement of Somali people in the project helped spur a positive trend in their relationship with the community at large (Lang et al., 2022), perhaps evidenced by the forging of new collaborations such as the local Good Neighbors Diversity Council.He also noted that the film in particular shed important light on the experiences of Somali youth, something perhaps overlooked within the Somali community itself.On the other hand, he expressed a wider concern that the film and conversations underrepresented the experiences of older firstgeneration immigrants, due in part to language and translation constraints.

Limitations and Future Directions
Evaluating community-based research and action projects demands sensitivity to context (Yardley, 2000) and requires negotiation and compromise.To accommodate our community partners' primary goal for the interventionto effect positive change in the communitywe had to balance the scope of our assessment against the burden our methods might place on collaborators and participants.We opted not to collect more extensive data and to collect all data as unobtrusively as possible.Conducting more interviews would enrich the data but also encroach on the county employees' ability to do the very work the intervention and this study were designed to support.Placing heavier demands also risked overtaxing and ultimately alienating our valued community partners.Therefore, we are also limited in how forcefully we can extrapolate findings from the interviews to the entire body of participants.
Although participants were interviewed by someone unrelated to the intervention's design and implementation, they might still have been reluctant to offer strongly negative feedback.The self-selected nature of the interview participants also opens the possibility that those with positive experiences are overrepresented.However, 1 of the 11 interviewees did express strong and overt resistance to some aspects of the film and panel.This suggests that our process did provide some level of invitation for participants to offer negative feedback.We also acknowledge the potential limitations inherent in having the first author participate in assessing the intervention he was invested in creating and implementing.We sought to offset potential bias by including the second author, an outside expert in assessment methodology who was not involved in the intervention design and implementation, in all stages of the analysis.
This study tracked the impact of intergroup interaction on the audience of the intervention: White, non-Muslim community members.It did not directly examine the impact on the Somali contributors to the film or discussion panels, or upon the Somali community at large.Though personal anecdotes suggest many Somali contributors found their involvement to be personally rewarding and beneficial, such input could be influenced by obligation or other factors that make it unreliable data.An important next step would be to study the effects upon the Somali community to determine whether and how the intervention has affected their lived experiences in our community.In order to maximize the potential of the intervention's positive effects on White attitudes-especially heightened reflexivity and cognitive dissonance-it would also be useful to engage with Somali residents and county officials to evaluate and revise policies and practices that have been harmful in the past.
The intervention under analysis shared several key characteristics with allyship training, setting it somewhat apart from the typical DEI training that emphasizes general concepts and expert insights.Our analysis indicated that this intervention affected participants in valuable ways, particularly related to perspective-taking, empathy development, and disruption of stereotypes.Such change is necessary to set a foundation for significant social change; without a critical mass of individuals who recognize and are motivated to act on social problems, broad-based change cannot be realized.However, interventions such as this cannot themselves motivate what Russell and Bohan (2016) label second-order change, the alteration of structures and hierarchies of power.While a community-based, outgroup-centered model for DEI intervention can counteract some of the pitfalls of common DEI efforts, it must be but one part of a broader, more explicitly critical approach to identifying and dismantling the assumed knowledge and practices that support structural forms of racism and oppression.
As researchers, we acknowledge that our positionality-how our own experiences and biases shape our perspectives relative to our research-can affect our observations in invisible ways.Feminist scholars and others remind us that unreflexive research risks reproducing oppressive ideas about marginalized peoples (Olesen, 2005).We embraced Finlay's ( 2008) approach for our work, interrogating the tension between equally unattainable goals of setting aside our predispositions to maintain objectivity and interrogating them fully to identify and offset our subjectivity.This allowed us to scrutinize our new understandings as they evolved dynamically throughout the research process.Some of our most salient positionalities include race (both authors are White), class (both rising middle class), education (both doctorate level), sex/gender (one cis maleidentified and one queer-identified), geography (both long-time residents of the area of study), political alignment (both liberal-progressive), prior connection to the object of study (one helped produce the film, other none), and familiarity with the study subjects (neither had personal or professional ties to the interview participants).

Applications for Community-Based DEI Work
As communities continue to grapple with the challenges of ever-diversifying populations, it is vital to assess and refine our methods for understanding the lived experiences of outgroup populations-for transcending fear and anxiety, and for ending harassment and violence that result from racism, anti-immigrant bias, and other forms of bigotry.The results of this study indicate the powerful contribution that communitybased collaborations can make in advancing diversity, equity, and inclusion goals.Specifically, interventions that include local experiences, focus on a specific population, and incorporate self-reflection were found to disrupt harmful stereotypes and foster personal connections.
First, community-based collaborations provided a platform for creating the social and parasocial contact important for forging empathy and countering bias.Our analysis highlights the importance for White people to make positive personal connections, even ephemeral or parasocial ones, with members of outgroup populations in their own communities.In majority-White areas, even willing White people have few opportunities for authentic interaction with underrepresented groups.A well-wrought community collaboration, particularly one that centers the perspectives and needs of the underrepresented group itself, can help address this problem.Participants felt strongly that a training centered on the inclusion of local people and grounded in local concerns heightened its impact by increasing engagement and setting a baseline for shared understanding.Given the higher investment that participants reported having in this intervention compared to didactic concept-focused formats, we conclude that one-size-fits-all content should be replaced by, or at least supplemented with, content created and delivered in collaboration with locals who have a stake in the well-being of the community.
Second, our findings indicate that focusing on a particular outgroup population (in this case, Muslim Somali immigrants) strengthened the impact of the intervention by disrupting specific stereotypes and inspiring more concrete reflexivity among participants.Although our analysis does not indicate whether or how the participants' shift in perception of one particular subgroup would translate to other subgroups, research has indicated that "single-group" training can be as effective as generic "inclusivity" training in shaping general attitudes (Bezrukova et al., 2016).Focusing on a specific subgroup in this instance heightened engagement-a necessary component to the success of any education effort-and has the potential to encourage more general patterns of inclusive thinking.
Third, we observe that while assessment plays a vital role in the ongoing review and refinement of both community-based learning and DEI training, assessment can also enhance the quality of the training directly.The interviews we conducted yielded new opportunities for participants to engage in reflexive thinking and perspectivetaking, both qualities of effective DEI training (Lindsey et al., 2015).Best practices also suggest that DEI education works best with consistent follow-up rather than in isolation (Bezrukova et al., 2016), and the interviews contributed in some measure to this follow-up as well.Researchers and trainers should also explore varied methods for engaging in the reflection and assessment process, for example through in-person or telephone interviews or through written or recorded journals.Questionnaires can also benefit the reflection and assessment process not only by providing data to researchers but also by providing prompts to spur participants to continue thinking about diversity, equity, and inclusion.
Finally, this project and others like it reflect a valuable, some argue necessary, shift in academic labor toward what Ernest Boyer dubbed "the scholarship of engagement" (2016, p. 11), dedicated to reconnecting academic resources to the pressing issues of our place and time but also to a broader mission to create "a special climate in which the academic and civic cultures com municate more continuously and more creatively with each other" (p.20).To this end, Boyer advocates expanding the traditional scholarly goals of discovering, integrating, and sharing knowledge to also include the application of knowledge (Boyer, 1990, p.13).Effective scholarship of application deploys a cyclical process moving from theory to practice and back again to theory: disciplinary knowledge and research methods are used to address real-world problems, then the outcomes of those practices are studied to improve the disciplinary knowledge base from which they draw.Our study attempts to model this theory-practice-theory cycle, demonstrating how theory-driven processes can inform effective community-engaged research and action.Thoughtful study of that action can then feed back into the collective knowledge base, through resources such as this very journal, to be redeployed for the future benefit of communities elsewhere.
(a) In what ways has the training influenced you, either personally or professionally?(b) How would you compare this training to other diversity trainings you've participated in?(c) Do you think this training had an impact?In what ways?(d) What could be done differently, either by the trainers or your organization, to improve the training outcomes?Respondents took as much time as And my mind went straight to [the screening and conversation].That really … helped me better understand where [Somali people] are coming from and what their culture means to them.(Participant 1)