In vivid colors and a stylish font, this urban atlas of the greater Boston area—itself part of a larger series of urban atlases that covered hundreds of other cities across the nation—plots statistical data gathered during the 1970 decennial Census. Starting in 1972, federal agencies including the U.S. Census partnered with the Lawrence Berkeley Laboratory (LBL) in Berkeley, California to design a computerized system that could make maps showing the distribution of demographic variables across space. It consisted of two major pieces of software: MAPEDIT and CARTE. First, using MAPEDIT, the Census Bureau digitized 35,000 standardized statistical areas called census tracts across 241 metropolitan regions. Then, using CARTE, they matched statistical data to its corresponding Census tract based on a common identifier. Today, this process of associating descriptive information about a place with geospatial data is commonly called a table join, and it’s a fundamental part of any digital mapping project. In the early 1970s, it was groundbreaking.
The distinctive racial geographies of the greater Boston area are on full display in the first and second of these maps, which document the stark differences between where Boston’s Black and non-Black residents lived. At the time of the 1970 census, Black Bostonians overwhelmingly lived along a corridor that connects Roxbury to Mattapan, roughly following the curves of Tremont Street, Columbus Avenue, Seaver Street, and Blue Hill Avenue. When this atlas was published in 1976, Boston was at the height of legal and social challenges around busing, a fraught effort to desegregate public schools that faced intense—and frequently violent—backlash from white Bostonians.
The third map, a bivaritate or “two variable” choropleth map, combines two different color ramps—pink and blue—to display information about both income and education at once. It seems to indicate that a high school diploma corresponds with a higher income, and vice versa. However, because census tracts are not very detailed units of areal measurement, we should always approach statistical maps like these with a healthy dose of skepticism. Rather than looking for clear-cut answers, you might use this map as a starting point to ask new questions about why different variables are clustered where they are.
In brilliant shades of blue and pink, the final map documents the irregular geographies of housing ownership in the greater Boston area. The deepest blues, which overwhelmingly cluster in more suburban places like West Roxbury and suburban towns like Dedham, indicate Census tracts where at least 90% of the housing units are occupied by someone who owns the property. By contrast, the deepest pinks indicate those tracts in which at least 90% of housing units are occupied otherwise (e.g., by someone who rents). Pink tracts consistently cluster in downtown Boston and Cambridge.