Background: "Activity space" has been used to examine how people's habitual movements interact with their environment, and can be used to examine accessibility to healthcare opportunities. Traditionally, the standard deviational ellipse (SDE), a Euclidean measure, has been used to represent activity space. We describe the construction and application of the SDE at one and two standard deviations, and three additional network-based measures of activity space using common tools in GIS: the road network buffer (RNB), the 30-minute standard travel time polygon (STT), and the relative travel time polygon (RTT). We compare the theoretical and methodological assumptions of each measure, and evaluate the measures by examining access to primary care services, using data from western North Carolina.
Results: Individual accessibility is defined as the availability of healthcare opportunities within that individual's activity space. Access is influenced by the shape and area of an individual's activity space, the spatial distribution of opportunities, and by the spatial structures that constrain and direct movement through space; the shape and area of the activity space is partly a product of how it is conceptualized and measured. Network-derived measures improve upon the SDE by incorporating the spatial structures (roads) that channel movement. The area of the STT is primarily influenced by the location of a respondent's residence within the road network hierarchy, with residents living near primary roads having the largest activity spaces. The RNB was most descriptive of actual opportunities and can be used to examine bypassing. The area of the RTT had the strongest correlation with a healthcare destination being located inside the activity space.
Conclusion: The availability of geospatial technologies and data create multiple options for representing and operationalizing the construct of activity space. Each approach has its strengths and limitations, and presents a different view of accessibility. While the choice of method ultimately lies in the research question, interpretation of results must consider the interrelated issues of method, representation, and application. Triangulation aids this interpretation and provides a more complete and nuanced understanding of accessibility.