Creating and Validating GIS Measures of Urban Design for Health Research

An objective system for assessing the characteristics of an urban neighborhood produced data comparable to that found using an observational method.

Urban design professionals describe city streets using several constructs: enclosure is what makes people feel bounded by buildings, trees and other vertical elements of a neighborhood; transparency refers to whether a passerby can perceive activity within buildings, through windows or fences. Other urban design constructs are imageability, human scale and complexity.

Studies have demonstrated an association between characteristics of city neighborhoods and physical activity. This research, however, is subject to limitations: studies that use direct observation are difficult to carry out on a large scale. Survey-based research relies on general descriptors like “attractive” or “interesting.”

This study tested a geographic information system (GIS) for the direct measurement of urban design features. Researchers surveyed 588 blocks in New York City using an existing observational protocol for measuring urban design. The authors used public data sources to derive GIS measurements equivalent to the measures in the observational protocol. For example, the observational protocol included a measure for the proportion of a street segment with windows. The GIS obtained this data from the NYC Department of City Planning department of land use tax output database.

Key Findings:

  • There were substantial correlations between GIS data and field observations.
  • The authors of this study found it impossible to construct GIS measures for all of the categories found in the observational protocol. Future research will likely face similar constraints.

This paper presents a methodology for developing objective, scalable measures of urban design. Because data sources are specific to every city, collaboration among researchers will be highly important.