This dataset contains 1,852 points that represent houses in the Soho neighborhood as featured in the 1855 map by the General Board of Health. For each house, the variables include an ID, the number of cholera deaths by residents and non-residents of the building (including non-deaths), a total death count for both types of residents, and a dummy variable that indicates whether the house was located on top of Craven Estate, a 17th-century pest-field. They also include three variables that measure distances to the pest-field, the nearest open sewer grate or ventilator, and the Broad Street pump.

The documentation is available at:


The Geoda script is available at:


Overview of data

Variable Description
ID Unique ID
deaths_r Number of deaths of residents per house
deaths_nr Number of deaths of non-residents per house
deaths Number of total deaths per house
pestfield Creates categories depending on whether the house is located on top of the former Craven Estate (1) or not (0).
dis_pestf Distance to the former Craven Estate, in meters.
dis_sewers Distance to the nearest open sewer grate or ventilator, in meters.
dis_bspump Distance to the Broad Street pump, in meters.
death_dum Dummy variables for deaths (1) or no deaths (0) in a house.
COORD_X X-coordinate in meters (see documentation for projection details).
COORD_Y Y-coordinate in meters (see documentation for projection details).

Prepared by Center for Spatial Data Science. Last updated February 1, 2021. Data provided “as is,” no warranties.