| Popis: |
To study how context influences health and well-being it is first necessary to measure the boundary of a person's geographical context. Rigorous description of context is difficult and existing studies have offered weak and mixed evidence about the importance of context to health. Either the characteristics of places have been inadequately conceptualized, or the spatial definition of place relies on inflexible administrative boundaries and a limited set of corresponding attribute data. Furthermore, the influence of context on health presumably depends a great deal on the amount of time spent in each space, and the potential for misclassifying context is greater for individuals whose routine activities are spatially diffuse. However, errors in spatial classification are usually not specified in studies linking context to health. This dissertation combines theories from geomatics and population health to argue that the time geography of human activities has an important role in delineating and describing contexts relevant to health. Specifically, the contextual impact on health status reflects not only the person's immediate neighbourhood, but also locations where other activities of daily life occur. The theoretical contribution of the thesis is to conceptualize context as a healthscape---a spatial notion that captures an array of contexts defined by individuals as they navigate through the spatial patterns of daily life. The empirical goal of the dissertation is to compare these healthscapes to contexts delineated by census tract geographies typically used in health and place research. Individual healthscapes are defined by developing a wearable global positioning system and data logger that records an individual's location and velocity at one-second intervals over a seven-day period. Data from 53 people are linked to responses from a health questionnaire. Several approaches drawn from ecology and spatial statistics are used to quantify the properties of each person's spatial activity patterns, or healthscape. These objectively measured contexts are compared statistically to the traditionally used census boundaries to examine differences in morphology, compositional and contextual characteristics. The results are broadly consistent with the theoretical justification proposed for healthscapes, and signal the need for place and health research to better represent the geography of human activities. |
