Drawing upon data derived from extensive written records as well as historical maps, this project investigates spatial patterns of residential segregation in Nineteenth Century American cities, exploring issues of wealth, occupation and race. More basically, every individual citizen, to the furthest extent possible, is mapped at his or her place of residence over the period of study in three major American cities. To accomplish this goal, the project team has developed a credible geospatial network, similar to the contemporary TIGER database, that covers Washington, D.C., Nashville, Tennessee in 1860 as well as Omaha, Nebraska for 1870. Address locator services were created based upon the historic street network, and census records represented as data points were matched with address information from city directories so they could be geocoded and adjusted to fit accurately to their true spatial locations. The research therefore has the effect of creating a resource for historical inquiry, and it enables immediate analysis of the study areas. When mapped, census records that contain information about personal estate values, real estate values, occupations and race provide quantitative spatial data about economic segregation and ethnic/racial segregation in three major American cities prior to the commercial introduction of the automobile.

Data are presented as one single layer on the map. Many people without GIS expertise will have much to contribute to these discussions, so the website has been created so that all visitors with basic Internet proficiency can access, filter and search through geocoded household-level records using this interactive application. The user interface provides all geocoded census records as points that, when clicked, generate pop-up windows with complete information about each head of household living at that location, along with the aggregated estate values for that individual’s family. Each person’s address or specific location from the city directories is also listed in the pop-up windows. Listing each address highlights the extent of differences between the historical street grid and the contemporary system, and although these data have been checked thoroughly for accuracy, it also provides a requisite level of transparency that will enable other experts to identify, document and notify the project of any possible issues with locations of individuals. A search tool on the top left of the screen allows visitors to type in a name to search for a particular individual. Using the power of a Fuse.js, an open JavaScript library, users can enter search parameters and get suggestions and similar results to what they are typing into the text-box in real-time with this tool. Even if site visitors do not know a particular name or its proper spelling in the census records, the search feature will calculate the most likely candidates based on the letters that have been entered, and put those results at the top of a list. Likewise, a basic radio button menu, displayed prominently on the upper right hand side of the screen, gives users the ability to filter data for various groups of interest to the particular research. Each city map has been populated with unique layers for people of color, immigrants from Ireland, immigrants from German territories and states, merchants of all types, laborers, and “elites” defined as bankers, lawyers and gentlemen. These categories are directly associated with the variables used for spatial comparison of various groups. When one of the categories in the filter is selected by a user, all of the dots with individuals who fit that specific criterion appear to highlight themselves immediately as a bright yellow layer that sits on top of the layer for all residents, providing a quick reference to all of their locations and allowing users to click directly on individuals in the group of interest, in order to obtain more information. Also, this feature allows site visitors to create custom map visualizations of the distribution of various groups. For example, a researcher inspecting the spatial distribution of people of color in Washington, D.C. can get a quick map of all of their locations in 1860 by clicking on the “People of Color” category.

In addition to making the interface simple enough for a broad variety of prospective users, the code for the site is also as basic as possible to help ensure digital preservation and long term compatibility with web browsers. Everything has been written with free and open source components, and created with simple HTML, CSS and JavaScript functions. Rather than storing the geographic data in a proprietary format or as map layer files that are native to specific GIS program, data layers are stored as text-based GeoJSON files. These are JavaScript Object Notation files, which tell web browsers how to interpret and use any attributes associated with each object, and they include an additional field to describe the geography of each object as points, lines or areas in basic Cartesian coordinates. Anybody with a web browser or text editor can view and access these GeoJSON files from this project directly as text files, edit them at will, and even upload them to their own web server if desired. They are listed in a downloads section for people who might want to do that. Furthermore, the more advanced scripts used in the map code itself are drawn from the popular open-source Leaflet.js, a JavaScript library designed for basic mapping applications.

For more information, contact:
Rob Shepard, project director