**5. Discussion**

The study results indicate there is a gap between domestic violence agencies and refugee communities in terms of familiarity. The results showed that by and large, executive directors were not able to name the most predominant refugee groups in their city. This is meaningful because the DV agencies surveyed were located in refugee resettlement cities, where refugee populations comprise a nontrivial portion of the wider population. When agencies lack even the basic knowledge of what refugee communities are most predominant in their city, they most likely lack more specific knowledge regarding cultural values and considerations within these communities that would improve their ability to e ffectively meet the needs of refugee DV survivors. For example, if an organization knew that one of the predominant refugee communities in their city was Somali, they might develop a basic understanding of Sunni Islamic traditions which could guide organizational outreach e fforts. This finding connects to the number of refugee clients served per year—where nearly half of reporting agencies indicated serving 10 or fewer refugees annually. Without enough exposure, agencies may not experience a sense of urgency to adapt their services to better serve refugee communities. However, it should be reiterated that the responding agencies were located in resettlement cities where overall refugee populations comprise a meaningful proportion of the wider population. Lack of agency exposure to refugee communities should not be equated with a lack of need within refugee communities

In line with a general lack of familiarity with the refugee communities in their city, executive directors generally reported lower frequencies of agency outreach e fforts to more informal refugee community contacts like faith leaders. The results suggested that, in general, DV agencies could improve interagency collaboration with refugee service providers as well as meaningful engagemen<sup>t</sup> with local refugee community leaders. In their 2009 report for the Family Violence Prevention Fund, Runner, Yoshihama, and Novick discuss how some DV organizations report little success in engaging existing community leaders in refugee and immigrant communities in the fight against domestic violence. Some of this may be due to a lack of organizational resources, and our results sugges<sup>t</sup> this may be due to a lack of organizational prioritization.

Although DV agencies reported less frequently hiring from refugee communities, they reported more frequently hiring bilingual sta ff and encouraging cultural competence training. Agencies generally reported frequently engaging in organizational practices that promote dialogue and critical introspection among their sta ff, but less frequently reported e fforts that promoted bidirectional exchanges with refugee communities. There seemed to be an inherent disconnect between how often agencies reported that they implement culturally responsive language practices for refugee communities and their agency budget for translation. While more than 8 out of 10 agencies reported that they use a translation service, almost 2 out of 3 agencies reported that they did not have an annual budget to support translation services. Agencies reported that they more often use formal translation services for help with refugee clients, as opposed to friends, family, and children; however, it is unclear how these services are being supported in agency budgets.

It is important to note that almost half (44%) of the agencies reported that 5% or more of their annual client population were from refugee communities in their city. When more than 5% of an agency's client base share an important identifying characteristic such as refugee status, it is critical that the agency develop and implement multi-level organizational approaches that support cultural responsiveness. Although among some of the weaker correlations, analyses showed that the percentage of clients served in the previous year who were refugees was positively associated with three of the six domains of cultural responsiveness, indicating that ongoing contact and cultural exchange can catapult organizations toward a deeper practice of cultural responsiveness.

Counterintuitively, the resources agencies allocate for translation services (one of the most critical supports according to the literature) was not associated with any of the domains of cultural responsiveness. Further, the length of time the agency had operated was inversely correlated with systems and processes for assessment and approaches to language concerns. This could reflect an entrenchment in prior ways of operating among older organizations. Older DV organizations are often better resourced financially and stand to make the biggest impact in their cities. However, if these same organizations are not incorporating culturally responsive organizational practices, refugee survivors will continue to pay the price. No other organizational structure variables were correlated with agency cultural responsiveness domains.

The associations between cultural responsiveness domains were very strong and predominantly positive (with the exception of program/shelter accommodations/adaptations). From agency values/practices to agency approaches to language concerns, the correlations among these domains ranged from 0.63 to 84. Specifically, the correlations between language concerns and other domains of cultural responsiveness were high. When agencies are actively taking steps to translate materials, provide translation services, and generally communicate with clients in respectful ways that prioritize their lived experience, it is not surprising that agency values and other domains of cultural responsiveness would be thriving. These findings may point to the idea that when agencies enact culturally responsive practices in one domain, they are more likely to enact them in another. These practices can build on each other and lead to further improvements. For example, when an agency commits to hiring bilingual and bicultural sta ff, the agency perspective will broaden and new areas for improvement will be identified and rectified. These practices can domino into a more culturally responsive system on the whole as each incremental improvement spurs the agency to take action in the next. Findings from the bivariate analyses underscore a resounding theme in the literature: that agencies must take *active* steps to build a more culturally responsive organization.

### *Limitations and Future Directions*

There are a number of limitations associated with this study. The first and foremost is the likelihood that agencies overestimated their cultural responsiveness due to social desirability and thus, the results may not accurately reflect the extent to which cultural competence and responsiveness permeates their agency. Secondly, the response rate of the survey (28%) likely points to selection bias and could further skew the results to overrepresenting the levels of cultural competence and responsiveness among domestic violence organizations. It is possible that organizations already invested in this work self-selected into the survey. Organizations that were more uncertain about their ability to serve refugee populations are more likely to have opted out of the survey, as Daniels and Belton (2015) found. Thus, the findings of this study cannot be generalized to all DV agencies in U.S. resettlement cities. Finally, due to the study's descriptive and exploratory nature, the results do not link culturally responsive strategies to specific outcomes for victims of domestic violence. The results give us a starting point for understanding the extent of cultural competence and responsiveness among domestic violence organizations; however, they do not o ffer clear guidance for how to prioritize organizational change to effect positive outcomes for refugee clients. This then, is a natural next for future research. Evaluating the e fficacy of specific culturally competent practices and organizational policies is critical in order to inform evidence-based practice with refugee domestic violence survivors.

**Author Contributions:** Conceptualization, J.L.L. and K.M.S.; methodology, J.L.L. and K.M.S.; formal analysis, J.L.L. writing—original draft preparation, J.L.L., K.M.S., and T.H.; writing—review and editing, J.L.L.; funding acquisition, J.L.L. and K.M.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded primarily by Fight Against Domestic Violence, with additional support from the Mountain West Center for Regional Studies and the College of Humanities and Social Sciences at Utah State University. The views presented in this paper do not necessarily reflect the views of the funders.

**Acknowledgments:** The authors would like to thank Anna Z. Blau and Laura M. Mora from the International Women's House for their expert consultation on instrumentation development. In addition the authors would like to thank the students from the 2018 graduating class of Utah State University's Master of Social Work program for their assistance with the sampling strategy.

**Conflicts of Interest:** The authors declare no conflict of interest.
