And it’s highly technical and very, very impressive. It took us a while to figure it out because it had never been done before, but it was worth it.
Methodology FAQ
NVI provides the data, tools, and support to residents and community organizations working to make their neighborhoods and the city a better place. It is a first-of-its-kind survey, designed with input from residents to measure and track progress on the issues people care about in their neighborhoods. NVI goes beyond providing data, convening conversations between neighborhood advocates and the community development organizations and policymakers making decisions about neighborhood investments.
NVI was developed by Detroiters for Detroiters and their allies. Community Development Advocates of Detroit partners with JFM Consulting Group and Data Driven Detroit to steward this community resource. NVI connects information from actual Detroit residents about schools, safety, parks, healthcare, and more, then combines it with public data to paint a picture of the strengths and challenges of Detroit neighborhoods.
There are six categories of indicators within the NVI. These categories group indicators into specific topic areas and help focus attention on specific aspects of neighborhood life.
The NVI incorporates both primary data (collected by the NVI survey) and secondary data (collected by other organizations and made available to the public). Primary data in the NVI generally reflects residents’ feelings about and experiences in their neighborhoods. Secondary data in the NVI comes from a variety of sources such as the City of Detroit, the American Community Survey, and the Wayne County Treasurer.
Primary data: The NVI survey is available to all Detroit residents 18 years or older. Survey respondents are required to provide a valid current address in the City of Detroit, and the addresses are used to both aggregate data to NVI Neighborhood Zones and City Council Districts, as well as to mail $25 gift cards to each respondent. Survey collection happens through multiple outreach initiatives including: flyers at high-traffic locations like laundromats, community development organization outreach, and City of Detroit email blasts.
Secondary data: Secondary data comes from a variety of sources. The American Community Survey (ACS) is the source for seven indicators. ACS data was downloaded from data.census.gov at either the census block group or census tract level, and aggregated to City Council Districts and NVI Neighborhood Zones. Data Driven Detroit also maintains address- and parcel-level datasets from the City of Detroit, Wayne County, Vericast, and other sources in Data Driven Detroit’s Integrated Property Data System (IPDS). IPDS data was extracted at the parcel level and aggregated to City Council Districts. Additional indicators derive from the State of Michigan’s Center for Educational Performance and Information, Michigan’s Department of Health and Human Services, and Michigan Traffic Crash Facts.
Indicator details are included in the data description (PDF).
When a survey collection period closes, the addresses provided by survey takers were filtered by JFM Consulting Group to identify responses that are real Detroit residents. For example, addresses that were from outside the city of Detroit are not included in the final dataset. Data Driven Detroit then takes the survey-level data and matches the responses to our address dataset which helps identify which neighborhood zone and council district the respondent lives in.
The NVI indicators are then calculated from the survey responses. The individual indicator calculations can be found in the data description (PDF).
When we process survey data for publication, all names, addresses, and any other information that identifies an individual is removed from the dataset. We store the original survey responses for future use in analyzing trends or providing custom cuts of aggregated, de-identified data, and this data is kept in an encrypted file store to protect it from unauthorized users. Data Driven Detroit tightly controls access to this environment and ensures adherence to industry best practices for securing sensitive data.
Council districts and neighborhood zones in Detroit range in size, number of parcels, and population count. These differences impact how we interpret data describing these places. For example, while Council Districts 2 and 5 have similar counts of people living below the poverty line, 26,520 and 26,415 respectively, their populations differ by nearly 20,000, meaning 26.1% of people living in Council District 2 and 31.7% of the population in Council District 5 are under the poverty line.
Read our full explanation of per capita
In designing the NVI, we had to answer the question: What geographic area is the right size? To answer this question, we convened the Defining Neighborhoods task force from our broad spectrum of partners. Their challenge was to create geographic areas within Detroit that were small enough to provide actionable data for community-based organizations, but large enough to protect individual privacy and give us a reasonable target for collecting a citywide survey. For privacy reasons, we never report data for individual names or home addresses. Instead, we report data for geographic areas that combine or "aggregate" all of the responses from that area.
Read our full explanation of Neighborhood Zones
