Think about data. Maybe you’re imagining a spreadsheet of numbers, or a stream of 1’s and 0’s. Now think about where that data lives. Is it on your computer or a company’s server? These are the two models of data ownership: either you own it or it is used by a large corporation. There are crucial problems with these models.
Corporations collect data useful for their business goals, and rarely do they think about local residents’ needs. But if you own data and store it on your computer, you probably aren’t able to aggregate enough data to draw conclusions about anyone other than yourself. A FitBit can tell you how many steps you take; however, it can tell you very little about the neighborhood you’ve walked in. That is where community databases come in. They show us a civically-minded alternative. They are also a lost chapter of Los Angeles County history.
Community databases emerged in the 1990s as personal computers became more affordable. Being able to analyze data at home and school, rather than through expensive business mainframes, was a big deal. Community residents and groups were consistently on the short end of the stick. They were caught in a bind, rarely with the equipment to crunch numbers. A 1996 article on The Democratization of Data captured the excitement of the time. “Community groups from low-income neighborhoods have the most to gain from full access to data,” David Sawicki and William Craig wrote, “yet the least capability to achieve that access.” They were trying to bring loop community groups back in.
The Long Beach Community Database sought to change that dynamic. It arose organically from ReThinking Long Beach, an organization founded by journalists, sociologists, and educators to help improve the lives of residents. Over a decade ago the group saw an opportunity to deliver high-quality data about the community through a website. They shaped their goals through meetings with residents at a local coffee shop. Community groups and residents described several needs. Non-profit organizations like the Gumbiner Foundation, an early supporter, needed data on neighborhood well-being. There also wasn’t yet a reliable source of data for the community. It could also help residents answer the question, “what’s happening in my neighborhood?” Citizens could then use the data to ask pointed questions of public officials to hold them accountable.
Co-founder Bill Crampon was passionate about drawing attention to persistent inequalities in Los Angeles county. He wasn’t shy about talking about his dedication to women and children. “The kids are getting screwed,” he said of the city’s educational system. He used his technical skills to aggregate and host data on education and public safety. The group’s latest report, the Long Beach Equity Atlas, used data to make the case for devoting more attention and funding to schools, particularly in African-American and Latino neighborhoods. Crampon and Alex Norman distilled the importance of a community database: “information is not a commodity of power, but a community resource for good.”
William Crampon presenting to Leadership Long Beach in 2015
Alex Norman, professor emeritus at UCLA, had long been Crampon’s partner in the database effort. He similarly describes it as serving the local community. “The Long Beach Community Database brought an analysis that was specific to the locality which the data depicted,” Norman said. They obtained data sets on education and public safety, combining them with demographic data. Each layer added sophistication to data analysis. “It included a cross-cultural analysis [and] added data that government sources were reluctant to share.” Crampon noted that local government conflates data with positive PR. “The problem with government databases is so often they only put in what hypes them.”
Alex Norman presenting to Leadership Long Beach in 2015
The Long Beach Community Database never shied away from showing the truth about what was going on in the neighborhood. They started with publicly available data sets like the American Community Survey. Others were paid for in trust and shoe-leather. For years Crampon had a handshake agreement with the Police Department. He would simply show up to obtain crime data sets. The data he was able to access revealed new insights when paired with related educational and economic data. It showed the persistence of crime in neighborhoods that received poorer education—a harmful spiral that the ACLU dubbed the “school to prison pipeline.”
Keeping the database online and up to date with the most recent data sets was a labor of love. Cleaning the data was time-intensive. Each large data set required careful cleaning to anonymize any personal information. Addresses proved particularly thorny since there was often no standard way in which locations were entered. One officer might have entered “348 Park Street” as the nearest address, while another might have entered two cross-streets. In an era before Google maps all the addresses needed to be standardized by hand. While this work often proved tedious, it benefitted a wide variety of residents and groups.
The Long Beach Community Database gave local organizations a platform to further their goals of improving local neighborhoods. For example, the youth organization Campfire used the data to generate baseline statistics to apply for grants that helped young boys and girls. Graduate students wrote dissertations using its data.
Unfortunately the Long Beach Community Database came to the end of its funding, in part due to competition. Since the mid-1990s community databases started networking, resulting in efforts like the National Neighborhood Indicators Project (NNIP). NNIP continues to pool efforts to create indexes to draw conclusions across geographic regions. More recently, “open data” portals run by the government made data available for analysis and download. Hype around “big data” rose, meaning community data wasn’t the only data in town.
Yet none of these successors—neighborhood indexes, open data, or big data—were exactly a “community database.” Despite their many community successes, Crampon talked about another reason why the Long Beach Community Database didn’t take off. He felt they didn’t connect enough with a wider public. Maybe their vision of an engaged, data-literate citizenry floundered because they didn’t have the capacity for outreach and education. Perhaps Long Beach, with its relatively modest population of half a million, simply didn’t have enough journalists, students, and non-profit organizations in the city interested in using data for civic purposes. Aggregating data from a larger geographic area than a single city would draw a bigger audience. Crampon is sanguine about the Long Beach Community Database, which he describes as “a fun, interesting fifteen years of my life.” He and Norman retired, although both remain active in community projects.
The idea of connecting data with improved community outcomes is more relevant than ever. The community still has a right to the data it produces. Journalists, students, and non-profit organizations use data to do work in Los Angeles neighborhoods. As Crampon said, using data requires more than just making data accessible and caring about the community. It requires education and outreach. But the benefits are there. Community databases foster distributive justice. We should think about what Anthony Townsend called “slow data” — “collected sparingly and by design, not harvested opportunistically from data exhaust.” Data does not have to be siphoned from unaware individuals and used for corporate goals. It can be ethically stored and collectively administered to meet local needs. Data should be a shared resource and community databases can help to fill that niche.