This post was authored by Allan M. Parnell, Ph.D., Cedar Grove Institute for Sustainable Communities.

We were asked by Reed Colfax of Relman and Dane, PLLC to determine whether there was an association between race and access to public water services in the Coal Run area in Muskingum County, Ohio. The Coal Run area is just outside of Zanesville, Ohio. In particular, Mr. Colfax was interested in having the patterns mapped. My colleagues in this analysis were Ben Marsh, Professor of Geography and Environmental Studies at Bucknell University, and Ann Moss Joyner, my colleague at the Cedar Grove Institute of Sustainable Communities. The results of our analysis were shown in a set of maps shown to the jury.

Census data clearly were not going to be useful because of the small size of the neighborhood. The core of the neighborhood is in two census blocks, but census data was not helpful because of the scale and distribution of residents within the blocks. Within each block, the northern part is predominantly white and the southern part is predominantly non-white. One plaintiff lived in a third census block, and the residents on the north side of Adamsville Road are in a fourth census block. There was no clear way to use Census data to show the whether race is a factor in access to public water. I proposed a house-by-house analysis within the neighborhood.

We obtained public Geographic Information Systems (GIS) data from Muskingum County. The key data were the parcel data, which identified all occupied houses in the study areas, the location of water lines with dates of construction, Zanesville’s city limits, and the street locations. Relman and Dane obtained water billing data giving the addresses of all houses with public water service. I field-checked the parcel data confirming that each property identified in the parcel data was an occupied house. We also geocoded the location of the water plant.

We knew the race of plaintiffs, but we did not know the race of the residents in the other houses. I designed a short survey asking the number of residents, the race of each resident and how long they had lived at that address. Under my supervision, two employees of Relman and Dane went door-to-door collecting the household data. The survey took place over two days. If one resident in a household was non-white, the house was coded as non-white. In the very few cases where we were unable to speak to any resident, the house was coded as unknown. Using the public GIS data, the household survey information, the plaintiff information and the addresses of houses with billed water service, Ben Marsh built the GIS layers for the maps showing the clear pattern of racial discrimination. I wrote the expert report using the maps, the survey information, and information from the plaintiffs.

Reed Colfax and John Relman decided to build the maps before the jury, layer by layer, adding information to the base map, explaining where the information in each layer came from.

Map 1 is the base map, showing the location of the Coal Run area relative to the city and the water plant and the roads.

Map 2 adds the location of public water lines as of the date when the case was filed.

Map 3 adds the location of occupied houses, showing the proximity of these houses to water lines. Note, however, that having a water line in front of your house does not necessarily mean that you have public water.

Map 4 shows which houses had billed water service. Note that one house south of the water lines had billed water service. This was an “special arrangement” made with a private line run to that house.

Map 5 introduces the race of the household with the water lines. The water line down Langen Lane clearly ends where non-white residences begin.

Map 6 adds billed water service again (the dark blue dots) confirming that most non-white houses did not have public water while most white houses did have water service.

While we believe that color schemes should be intuitively obvious (and thus white and black make sense here), it is difficult to use true black and have any internal symbol (the dark blue dot meaning billed water service) show up well. Thus, to designate houses of minority residents, a dark orange or light brown might have been more informative. Regarding the maps as shown, the attorneys chose the color scheme of the houses to indicate race.

A final map shows how far the public water lines extended. Note the location of the Coal Run area in blue.

The defense used GIS to try and make the case that race did not affect who had water service and that some areas with African American residents had water service so there was no pattern of discrimination.

The defense map pictured left shows the Coal Run area and surrounding area. Census blocks are coded by the number of African American residents. Water lines are shown in blue. The defense expert argued that all residents of a census block had water if a water line intersected any part of the block. This is demonstrably false, and he had difficulty with this argument in his testimony. The defense expert also pointed to the census block southwest of the Coal Run area that borders I-70 with 34 African American residents.

I had examined the census data for the block and found that while there were 34 African American residents in 2000, there were no African American households. Individuals in the census are classified as living either in households or in group quarters, and the African American residents all lived in group quarters. And they were all elderly. Clearly, they live in a nursing home that is 88% white. The water line in question services a health care facility that had no residents (black or white) when it was built.

This post was authored by Allan M. Parnell, Ph.D., Cedar Grove Institute for Sustainable Communities.

Webmaster’s note: Plaintiffs in this case received a near 11 million dollar jury verdict (attorneys fees reserved).