Perceptions of Policy and Practice in Five Cities
Patricia Cobar, Kathleen A. Tomberg, Fidel L. Osorio and Jeffrey A. Butts
September 2024
Executive Summary
The John Jay College Research and Evaluation Center and the Urban Institute helped the National League of Cities’ Center for Municipal Practice to support local government leaders in five U.S. cities as they worked to reimagine their visions for public safety. The Municipalities Reimagining Community Safety (MRCS) initiative was supported by the Wells Fargo Foundation’s Social Impact and Sustainability commitment.
As research partner for the project, the Research and Evaluation Center (JohnJayREC) designed a three-wave survey focused on measuring the perceptions of local stakeholders involved in public safety efforts in each city. Results from the survey highlighted stakeholder opinions about the efficacy and equity of their city’s approaches to crime prevention and community well-being. Researchers explored trends across and within cities to assess the effects and success of the MRCS initiative.
Introduction
The John Jay College of Criminal Justice Research and Evaluation Center (JohnJayREC) worked in partnership with Urban Institute (Urban) to help the National League of Cities (NLC) Center for Municipal Practice support five cities working to build stronger, safer communities (Baltimore, MD; Dallas, TX; Jackson, MS; Philadelphia, PA; and St. Louis, MO). The Municipalities Reimagining Community Safety (MRCS) initiative supported by Wells Fargo’s Social Impact and Sustainability commitment was designed to provide implementation funding, evaluation, and technical assistance resources to each city to enhance community-centered public safety innovations.
During the MRCS initiative, local elected leaders, city staff, residents, and local organizations convened to pursue coordinated, equitable, and community-centered public safety efforts. Each city used project resources to advance comprehensive safety/violence prevention efforts and to support the development, implementation, or enhancement of programs and policies matched with local violence prevention priorities.
NLC and Urban Institute led technical assistance efforts for MRCS projects. Throughout the initiative, NLC offered sites a series of peer learning meetings on various topics, such as best practices for school-based and hospital-based interventions. During these meetings, leaders from the participating cities convened with guests who were experts in varying fields to discuss their city’s planning and needs. Other topics addressed in these meetings included data acquisition and sustaining the MRCS initiatives after the funding period. Local groups worked collaboratively to design implementation strategies and follow up as needed. NLC and Urban Institute supported each city throughout the project and, based on need, provided resources like templates for data-sharing agreements, draft position descriptions for local hiring, community engagement planning, and shared potential partners for each team’s consideration.
NLC and Urban Institute conducted site visits, organized in-person focus groups, and provided technical assistance and real-time support to the initiative’s needs on the ground. Focus groups included individuals affected by violence in their communities to inform local conversations at the request of each MRCS site. Project findings were shared with site teams intermittently to provide policy insights and to enhance the array of efforts during the project. In preparation meetings for the site visits, NLC asked the local initiative team members what specific resources NLC or Urban Institute could provide as part of their visits. Such assistance would likely include literature reviews on relevant topics, the development of agency organizational charts, and consultation with outside experts.
JohnJayREC led efforts to measure MRCS city stakeholder opinions on questions related to public safety and violence reduction efforts. Survey findings revealed opinions and perspectives on the challenging work of reducing and preventing crime and violence, organizing partner networks across various sectors, collaborating with government and non-profit organizations, and sharing resources for community residents. Some cities reported relatively stable opinions across the 12 months of survey administration, while others reported shifting opinions (both negative and positive).
City Efforts
NLC supported efforts by local governments in Baltimore, Dallas, Jackson, Philadelphia, and St. Louis to develop comprehensive, community-centered violence prevention/intervention strategies. The cities varied in size, government structure, and in the phases of their efforts to improve the equity and impact of violence prevention and intervention strategies.
During the course of the MRCS initiative, each city received $700,000 to advance strategies of their choice. Each city proposed initiatives in response to the unique circumstances of their city and plans ranged from school-based violence prevention programs to launching city offices of violence prevention. The project ran from June 2022 to September 2024, which included a three-month extension period. A delay in onboarding paperwork meant some sites were unable to begin project activities until a few months into the project.
Baltimore, Maryland
The Baltimore Mayor’s Office on Neighborhood Safety and Engagement developed a school-based violence intervention program for neighborhood youth. The city planned to pilot the program during the 2024-2025 school year in at least three local high schools: Mergenthaler High School, George Washington Carver Vocational Technical High School, and Digital Harbor High School. A partner agency provided implementation support for the initiative, including designating staff for schools to run daily programming. Baltimore’s priorities were violence prevention and community safety with additional support from local partners and the public school system. The school-based initiative complemented existing community violence interventions to support youth most at risk of engaging in violence or being harmed themselves.
Dallas, Texas
The City of Dallas partnered with Lone Star Justice Alliance and other community organizations to focus efforts on enhancing services for youth and young adults affected by violence, and supporting their families to reduce violence and justice involvement. The multi-part intervention included the creation of the Youth Safety Collaborative, which built the city’s capacity to work with community organizations, and developing a data dashboard for youth-serving organizations that compiled metrics related to safety. The dashboard included data on youth crime, education, 311, and employment indicators.
Jackson, Mississippi
City leaders created an Office of Violence Prevention and Trauma Recovery (OVPTR) to support community and hospital-based violence interventions with a trauma-informed, public health approach. The initiative’s efforts helped build out the office’s capacity to serve Jackson residents and allowed it to hire a new director who conducted multiple outreach efforts to hear from community residents and provide strategic support following varied crises in the city, including violence. Key investments were launched by the summer of 2024.
Philadelphia, Pennsylvania
The city’s Office of Public Safety, with support from Philadelphia’s Department of Public Health and other local partners, relied in part on support from the MRCS initiative to implement a comprehensive approach to violence reduction. The city helped residents to improve community safety through various programs and initiatives that were designed to provide behavioral health interventions, job training, violence interruptions, and other opportunities in areas affected by violence. The effort aimed to support hospital-based violence interventions, enhance the wellness and capacity of community-based violence intervention staff with group retreats, and to build staff credibility by developing an accredited certification for the community intervention specialists known as “credible messengers.”
St. Louis, Missouri
St. Louis created the Office of Violence Prevention to support citywide efforts to reduce violent crime. The office was to be equipped with staff who could prioritize public health approaches to violence reduction and work alongside the professionals of the city’s Department of Health. The office supported alternative 911 emergency responses, community violence interventions, community reentry, youth diversion programs, and community outreach events that could convene violence prevention partners and residents and provide insightful discussion and supportive interactions with the broader community.
City Surveys
As one component of its efforts to assess the success of MRCS, the National League of Cities asked JohnJayREC to serve as an evaluation partner by measuring local perceptions of each city’s efforts and to explore awareness and comprehension of the MRCS initiative through a stakeholder survey. Researchers at JohnJayREC worked with NLC to design a survey that asked local stakeholders about their anti-violence efforts and the activities and possible impact of the MRCS initiative in each community. The survey was administered three times between March 2023 and March 2024, with approximately six months between each of the three waves. (Appendix A)
Surveys took approximately 10 minutes to complete online. After some basic demographics (e.g., age, education, employment field, etc.), the survey asked 42 Likert-scale questions, each designed to be clear, simple, and quick to answer. The topics included components of public safety, opinions on local violence reduction efforts, and contributions towards public safety by government and non-profit groups. The items aligned with different social factors that may influence public safety – community engagement, community services, criminal justice policies, economic influences, family services, social disparities, government priorities, resident collaboration, and resource sharing. To create the survey, the research team reviewed validated scales and previous experiences with other JohnJayREC surveys. Survey items were adapted as needed.
Survey Respondents
JohnJayREC, Urban, NLC, and the MRCS core team members in each site collaborated to compile a list of at least 40 potential survey respondents in each city, concentrating on key stakeholders in the local government and the nonprofit sector. Ideal survey respondents would know about violence in their neighborhoods and be involved in local violence prevention efforts.
Respondents were often the leaders and directors of local agencies and offices, project managers, senior line staff, and other community members who were involved in public safety efforts. After each list was drafted, the JohnJayREC team vetted the names to ensure key stakeholders were included. Final respondent pools included a mix of people who were and were not aware of their city’s MRCS efforts, but all of whom were knowledgeable about the development of public safety efforts in the city.
Researchers emailed each respondent a unique ID number and a link to the survey that was specific to their city. As necessary, the research team followed up with a series of three reminders (approximately one week apart) encouraging stakeholders who had yet to respond that their views and opinions were essential for ongoing public safety discussions. Researchers tracked individual survey completions using the ID numbers. Email reminders stopped once a respondent completed the survey.
In addition to individual reminders, the research team asked each city’s MRCS leadership members to send short endorsement emails to potential respondents at the start of each survey wave and before the issuance of any reminders from JohnJayREC. Support emails from the MRCS core team boosted participation in the survey. The emails introduced and emphasized the importance of the survey, boosted survey legitimacy, and primed stakeholders to look out for JohnJayREC’s next survey invitation.
The overall response rate to the surveys averaged 36 percent across all cities and all three waves (Table 1). The total response rate varied from a low of 30 percent in Baltimore to a high of 43 percent in Dallas. Those responding to the survey were employed in a variety of fields. Most respondents across all cities were either officials and staff from state or local governments (39%) or community agencies and the nonprofit sector (35%). Other respondents were from law enforcement (8%) or allied fields including education (5%), health care (3%), the faith community (2%), social services (1%), and private business (2%).
The employment sectors of survey respondents varied between participating cities (Figure 1). While survey respondents in Baltimore and Dallas were evenly distributed across the major categories of state and local government, nonprofits, and “other,” the remaining three cities had more divergent respondent pools. Survey respondents in Jackson were mostly from nonprofit and community agencies (63%). Most Philadelphia respondents were from state or local government agencies (65%), while respondents in St. Louis were least likely to be employed by government agencies (18%) and were more evenly divided among nonprofits (45%) and other affiliations (37%).
Respondents were highly educated. Across the three survey waves in all five cities, nearly 20 percent of respondents (19%) had earned a four-year college degree at the time of survey completion, and almost two-thirds (66%) reported either some post-graduate work or a completed graduate degree. Across all three waves, the percentage of respondents with a graduate degree or at least some graduate work varied from a low of 49 percent in Dallas to a high of 79 percent in St. Louis (Figure 2).
Findings
The survey first asked respondents to indicate their agreement with various resources that could be considered important for creating community safety. Across all five cities, respondents were asked to agree or disagree that community safety is supported by factors such as police and criminal justice agencies, housing stability, mental health supports, drug and alcohol treatment, schools, social services, etc.
For each type of resource, respondents answered either “disagree strongly,” “disagree somewhat,” “agree somewhat,” “agree strongly,” or “neutral.” If a respondent left a question blank or indicated that they preferred not to answer, their responses are omitted from the results. Researchers scored each item according to the strength of agreement and then ranked all 16 items in each city to examine the degree to which respondents differed in their assessments of various resources (Table 2).
The most consistent and positive agreement across all cities focused on economic opportunities, housing, schools, and community support. Items related to economic influences (e.g., jobs and income stability, neighborhood economic stability, and housing stability) received the highest extent of endorsement in all cities. Criminal justice and law enforcement’s contributions to public safety received the lowest endorsements in four of five cities. Dallas respondents, however, ranked policing fourth overall.
Respondents were next asked about their perceptions of local government leaders in their cities and how those officials approached community safety (Figure 3). Their views differed somewhat between cities and some responses changed between the first and third surveys (i.e., March 2023 to March 2024). Such changes could be due to shifts in perceptions of city leaders, or to changes in the composition of survey respondents, or perhaps both.
Perhaps the most fundamental question in the survey asked respondents whether they agreed or disagreed with the statement: “Local government leaders in [this city] know what to do to prevent crime and violence.” In the first wave of surveys in 2023, most respondents either disagreed with the statement or expressed no opinion and remained “neutral.” The percentage of respondents failing to agree with the statement ranged from 59 percent in Philadelphia and St. Louis to 68 percent in Jackson. With the exception of Dallas, the proportion of respondents failing to agree remained largely unchanged or grew slightly by 2024. In Dallas, more respondents agreed with the statement in 2024 than in 2023 (70% versus 39%).
Respondents were asked to agree or disagree that it was “hard to make real progress in reducing crime and violence.” Stakeholders in Baltimore and Philadelphia were inclined to agree. In 2023, 61 percent of Baltimore respondents either agreed somewhat or agreed strongly with the statement. One year later, the total of somewhat and strongly agree has grown to 75 percent. This could reflect an increased concern about the challenges facing communities. If so, respondents in St. Louis were even more concerned. Total agreement with the statement among St. Louis respondents grew from just 37 percent in 2023 to 80 percent in 2024. On the other hand, respondents in Dallas reported less agreement in 2024 than in 2023 (45% versus 56%).
Other differences were apparent when comparing cities. When respondents were asked if local government leaders “cared about our neighborhoods,” the answers in 2023 were quite positive in Dallas (88% either somewhat or strongly agreed). Similar positive responses in 2023 came from Baltimore (82%), Philadelphia (77%), and St. Louis (82%). Jackson was slightly less positive, as just 58 percent of respondents agreed with the statement, either strongly (21%) or somewhat (37%). The perception of Jackson respondents, however, improved between 2023 and 2024, growing from 58 percent to 63 percent in total agreement.
In the other four cities, 2024 respondents were less likely than those in 2023 to agree that local government leaders cared about their city’s neighborhoods. In other words, they were more likely to disagree or to report no opinion on the question. The percentage of respondents indicating more disagreement increased in Baltimore (from 18% to 31%), Dallas (12% to 25%), Philadelphia (23% to 41%), and St. Louis (18% to 26%).
Similar patterns were observed for other survey items. The survey asked respondents to assess whether local government leaders “care about the needs of residents,” and the cities again revealed different patterns. Two-thirds to three-quarters of the respondents in Baltimore, Jackson, and St. Louis agreed with the statement, and their responses did not change markedly between 2023 and 2024. In Dallas, however, overall agreement grew (67% to 80%). In Philadelphia, total agreement to the same statement fell from 84 percent to 62 percent between the first and third survey.
In Philadelphia and St. Louis, survey respondents were least likely to agree that local government leaders “listen to residents about public safety.” The proportion agreeing, either somewhat or strongly, fell from 66 percent to 42 percent in one year in Philadelphia and from 69 to 56 percent in St. Louis. In the other three cities, the proportions of respondents agreeing with the statement actually grew slightly between 2023 and 2024.
The survey also asked respondents if local government leaders “listen to experts about public safety.” Majorities in Baltimore, Dallas, and St. Louis agreed with the statement in 2023, but their responses were slightly less positive in 2024. This was especially true in St. Louis, where the proportion of respondents agreeing fell from 75 to 50 percent. Philadelphia was the only city where respondents were more likely in 2024 than in 2023 to agree that their city leaders listened to experts about community safety issues. The proportion of Philadelphia respondents agreeing with the statement increased from 48 to 58 percent.
Respondents in all five MRCS cities reported high levels of agreement that their local government leaders cared about the “needs of law enforcement.” In 2023, nearly all respondents in St. Louis either agreed somewhat (69%) or agreed strongly (25%) with the statement, for total agreement of 94 percent. Dallas respondents were also very likely to agree (89%). In all five cities, however, respondents in 2023 were less likely to express agreement with the statement.
The survey included two questions designed to elicit the views of any respondents who may have been harboring negative opinions about the sincerity and judgment of local officials. One question asked for agreement or disagreement with the statement “local government leaders in [this city] care more about some neighborhoods than others.” Respondents in St. Louis were most likely to agree in both 2023 (88%) and 2024 (84%). Baltimore respondents were least likely to agree in 2023 (59%), and those in Jackson were least likely to agree in 2024 (55%). In Baltimore and Philadelphia, respondents agreed with the statement slightly more in 2024 than in 2023, while those in the other three cities were somewhat less likely to agree.
Another provocative question asked respondents whether “local government leaders in [this city] mainly care about their own careers.” In three of the MRCS cities, the proportion agreeing with the statement grew between 2023 and 2024, either slightly (from 40% to 46% in Baltimore) or more substantially (26% to 61% in Philadelphia; 29% to 61% in St. Louis). In the other cities, respondents were less likely to agree with the statement in 2024 than in 2023. Total agreement (either somewhat or strongly) fell from 50 percent to 42 percent in Dallas and from 74 to 50 percent in Jackson. Again, the survey samples were small and not necessarily representative year-to-year, so these changes may indicate greater support for local leaders or perhaps the different respondent pools in each city.
Respondents were asked to assess the diversity of community safety approaches in their cities, and which communities were most likely to be harmed by crime and violence (Figure 4). Not surprisingly, large majorities of respondents agreed that public safety problems mostly harm “communities of color” and “poor communities.” Of note, however, are the shifts toward lower support for these statements between 2023 and 2024, especially in Dallas and Jackson.
When asked about their local government’s collaboration with diverse groups and organizations, respondents tended to be more supportive in Baltimore and Dallas than in Jackson and Philadelphia. In Jackson, for example, half (50%) of respondents in 2023 and more than half (55%) in 2024 either disagreed or offered no opinion as to whether their public safety efforts involved “diverse groups of people and organizations.” In Dallas, on the other hand, nearly all respondents agreed with the statement. Their agreement (somewhat plus strongly) totaled 89 percent in 2023 and 80 percent in 2024.
Survey respondents were asked to indicate their assessment of the degree of support provided by local governments and nonprofit agencies in supporting public safety (Figure 5). Their opinions were generally strongest in reacting to the support received from nonprofits, and some responses improved over time.
In Baltimore, for example, survey respondents agreed somewhat (38%) or agreed strongly (10%) that local government leaders supported public safety by sharing financial resources with neighborhood residents. One year later in 2024, their responses grew to 35 percent and 39 percent, respectively. Most of the cities responded similarly, with the exception of Jackson. Respondents there were very favorable about the support received from the nonprofit sector (72% total agreement in 2024), but far less positive about the support of local government leaders (31% total agreement in 2024).
Respondents were asked to evaluate the degree to which local governments and nonprofit agencies supported public safety by sharing information and data with neighborhood residents (Figure 6). In most of the MRCS cities, the nonprofit sector was rated more highly. Respondents in Dallas were less positive in 2024 than in 2023 for local government leaders. Jackson respondents rated the support of nonprofit agencies far more positively than the support of local government. Total agreement about the support of nonprofit agencies in Jackson was 81 percent in 2023 and 100 percent in 2024. While respondents in St. Louis were also generally favorable about the data supports from nonprofits, their support declined from a total of 83 percent to 73 percent between 2023 and 2024. Total agreement for the support of local government grew from 53 to 69 percent.
Respondents then rated the support for public safety that was generated by local governments and nonprofit agencies in collaboration with neighborhood residents (Figure 7). The responses were generally positive and often increased between 2023 and 2024. In Philadelphia, for example, total agreement (somewhat and strongly) about support from local government leaders grew from 50 percent to 69 percent between 2023 and 2024. Total agreement of Philadelphia respondents about support of the nonprofit sector grew from 75 percent to 80 percent. By 2024, the other cities were also very positive about the support created from collaborations of residents with nonprofit agencies (88% total agreement in Baltimore and Dallas, 86% in St. Louis, and 100% in Jackson).
Conclusion
The NLC Municipalities Reimagining Community Safety Initiative supported community safety efforts in Baltimore, Dallas, Jackson, Philadelphia, and St. Louis. The JohnJayREC surveys were designed to inform MRCS initiative leaders about local perceptions of public safety policies and practices in each city. Survey respondents highlighted many differences in support received from local government leaders and nonprofit organizations, and they generally agreed that local nonprofits supported public safety effectively through resident collaboration and resource sharing.
Opinions on the support of local governments often varied between cities. Differences may be due to the array and complexity of the procedures and bureaucracy local governments must maneuver to support their constituents as well as the focus of each city’s effort. Local governments may intend to provide rapid and targeted support for neighborhoods and residents, but it could be challenging to administer resources and focus them effectively on the areas most in need.
Each city involved in the MRCS initiative was given the flexibility to determine the mission and scope of the initiative and to shape strategies to help the community. The research team accommodated these differences by designing the survey to record general perceptions of public safety concerns in ways that could be informative for all five cities. The results suggest that stakeholders in all sites agreed that economic supports such as employment, income, and housing stability were the most effective means of creating public safety. Stakeholders across the MRCS cities also agreed that public safety problems disproportionately affect communities of color and poor communities.
Findings from the JohnJayREC survey provide valuable information about public safety efforts in each MRCS site, but the survey results are not meant to stand alone. For a complete understanding of the initiative and its effects, NLC also commissioned Urban Institute to conduct a series of focus groups to enable local leaders and community residents to offer observations of MRCS more directly. Additional information about the MRCS initiative is available from the National League of Cities.
Funding
The National League of Cities and Wells Fargo Foundation provided funding support for this analysis, but all conclusions are those of the authors. Funders and partners of the Research and Evaluation Center are not responsible for any findings presented in Center publications.
Acknowledgments
The authors appreciate the contributions of other staff members at the Research and Evaluation Center who organized this project and reviewed drafts of this report. In particular, we appreciate the assistance of Rebecca Balletto and Rhoda Ramdeen. The authors are also indebted to the staff of NLC and the Urban Institute who met with the research team regularly to share knowledge and experience. Their contributions helped to shape and inform the final product.
Recommended Citation
Cobar, Patricia, Kathleen A. Tomberg, Fidel L. Osorio and Jeffrey A. Butts (2024). Reimagining Community Safety — Perceptions of Policy and Practice in Five Cities. New York, NY: Research and Evaluation Center, John Jay College of Criminal Justice, City University of New York.
