Uncertain Geographic Context Problem: Implications for Environmental Health Research

The modifiable areal unit problem (MAUP) is a fundamental issue much of environmental health research has faced. Studies that examine the effects of area-based environmental attributes on individual health behaviors or outcomes also face another fundamental methodological problem. This is the problem that findings about the effects of area-based attributes may be affected by how contextual units (e.g., neighborhoods) are geographically delineated and the extent to which these areal units deviate from the “true causally relevant” geographic context. It arises because of the spatial uncertainty in the actual areas that exert the relevant contextual influences and the temporal uncertainty in the timing and duration individuals are exposed to these influences. The problem is referred to as the uncertain geographic context problem (UGCoP), which is a significant methodological problem in environmental health research because it means that research results can be different when different delineations of contextual units are used (as observed in several recent studies). The UGCoP is a problem as fundamental as the MAUP for any study that uses area-based contextual attributes. But it is a different kind of problem because it is not due to the use of different zonal schemes or spatial scales for area-based variables. Dynamic conceptualizations of context and new analytical methods that take human mobility into account are needed to address the UGCoP. In this presentation I discuss the nature and sources of the UGCoP. Using recent studies on neighborhood effects and environmental health as examples, I discuss how GIS-based methods and high-resolution space-time data collected with location-aware devices (e.g., GPS) may help mitigate the problem in health research.