Reducing Violence in Communities: An In-Depth Study of Efforts in Durham, NC and Minneapolis, MN, 2010-2022 (ICPSR 38691)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Hide

Study Purpose View help for Study Purpose

The research team aimed to answer the following questions:

1. What are community level factors that can contribute to persistent violence?
2. What are the key factors in both cities that distinguish high violent crime areas compared to low violent crime areas?

Study Design View help for Study Design

A mixed methods study design was used to address the research questions.

Key informant interviews. In order to gain knowledge about violence in both sites, the research team conducted semi-structured interviews with 27 local stakeholders involved in running anti-violence programs and other relevant programs. Interviews were recorded, with notes taken simultaneously. Data were coded using Dedoose by a member of the research team under the supervision of a senior qualitative researcher.

Qualtrics survey. To gain resident perceptions of violence across both sites, the team distributed an online survey via Qualtrics between June 16 and July 18, 2022. Qualtrics panel members in the Durham and Minneapolis metropolitan areas were notified that they were a candidate for this study and were invited to participate.

Administrative data. Police incident data were requested from Durham Police Department and Minneapolis Police Department to calculate violent crime rates by census block group. Neighborhood demographic and economic measures were sourced from American Community Survey (ACS) 5-year estimates at the census block group level. Commuting network data were calculated using Origin-Destination Employment Statistics (LODES) data from the Longitudinal Employer-Household Dynamics (LEHD) program at the U.S. Census Bureau. LODES data were aggregated from the block level to census block groups for analysis.

Sample View help for Sample

The two sites selected for this study, Durham and Minneapolis, are medium-sized cities in the United States that, at the time of study, had relatively high violent crime rates considering their population size, were comparable to many other medium-sized U.S. cities based on demographics, socioeconomic climate, and population size, and had both implemented anti-violence efforts through different avenues (e.g., law enforcement, local government, federal government, nonprofit organizations).

Key informant interviews. Initial interviewees were selected among local stakeholders of anti-violence efforts, including government, law enforcement, agency representatives, and members of neighborhood organizations and local non-profits. Snowball sampling was then used to identify other potential participants who had knowledge of violence reduction efforts in each site.

Qualtrics survey. Respondents for the community survey were drawn from Qualtrics panel participants. If not already on the panel, potential respondents completed a recruitment survey to qualify and be added to the panel. To be included, individuals were at least 18 years old, resided in a zip code within Durham or Minneapolis, and had not worked for law enforcement for at least 1 year prior to taking the survey.

Time Method View help for Time Method

Universe View help for Universe

• Census block groups within Durham, North Carolina and Minneapolis, Minnesota.
• Adults living either in Durham or Minneapolis at the time of study.
• Stakeholders involved in or with knowledge of anti-violence efforts in Durham or Minneapolis.

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Description of Variables View help for Description of Variables

Several measurements were calculated based on police incident data, American Community Survey (ACS), and Census Origin-Destination Employment Statistics (LODES) data from 2009-2018:

• Violent crime rate by census block group: Violent crime includes aggravated assault, robbery, homicide, and (for Durham only) sexual assault.
• Concentrated disadvantage: For this item, ACS component measures used were percent of households below 100 percent of the poverty line, households with public assistance income, single female head of household, unemployment, and resident under age 18. Z-scores of each component were used to create the index. Higher values indicate more disadvantaged block groups.
• Commuting network concentrated disadvantage: For each block group in a city, block group pairs were created with residents' homes and where they commuted to work. A weight was constructed that was the number of residents from the home block group commuting to the work block group, divided by the total number of commuters to all work areas from that home block group. The network index is the weighted sum of state-wide concentrated disadvantage scores for each connected block group.
• Racial heterogeneity per block group: 1 minus the sum of the squared proportion of residents in four racial categories (non-Hispanic White, non-Hispanic Black, Hispanic, all other races). Higher values indicate greater heterogeneity.
• Neighborhood characteristics: ACS components used were percent of renter-occupied housing units, vacant housing units, and overcrowded units; age of housing stock, resident mobility, rental cost, and property values.
• Gentrification indicators: Block groups were considered gentrified if the median household income was below the city-wide median, if more housing units were built before 1990 than the city-wide median, if the change in the number of college graduates since 2010 was greater than the median city-wide change, and if median property values increased in the block group. Block groups meeting the first two criteria only were considered eligible to gentrify.

Please refer to the available P.I. documentation for further details on secondary data sources and construction of derived variables.

Participants in the Durham and Minneapolis community surveys answered items in the following categories:

• Individual demographics: employment status, sex, race, Hispanic ethnicity, age, household income, and education level
• Interactions with police: number of times contacted by or contacted police in the last 2 years, if participant or close friend/family had been harassed by police
• Perceptions of police: desired level of community involvement, police budget, fairness, quality of services, trustworthiness
• Neighborhood cohesion and safety, including involvement in any neighborhood community groups

Participants in key informant interviews were first asked about their broader community context, specifically crime and violence present within the community, perceptions of why it is an issue, factors that lead to persistence, and barriers to solving the issue. If associated with anti-violence efforts, participants were also asked questions about these efforts: population served, types of activities conducted, funding sources, program/initiative goals, and persistent barriers.

Hide

Original Release Date View help for Original Release Date

2024-10-30

Version History View help for Version History

Hide