Mylène Riva

Au doctorat

Intérêts de recherche : Santé publique, Promotion de la santé, Géographie de la santé, Inégalités sociales de santé, Déterminants socio-environnementaux de la santé, Activité physique, Vie active, Surpoids et obésité, Analyses multiniveaux, Systèmes d’information géographique, Évaluation des incidences sur la santé (Health Impact Assessment)

Projet de thèse ou mémoire en cours

«Conceptual and Methodological Issues in Operationalising the Neighbourhood: The Example of Walking»

When operationally defining spatial units in order to understand area effects on health, several conceptual and methodological issues must be addressed including: 1) the modifiable areal unit problem (MAUP) or the fact that analytical results are sensitive to the definition of spatial units at which data are aggregated, and 2) the homogeneity of areas or the uniformity of the ecological dimensions along which small areas should be conceptually and operationally defined.

To address these issues, this thesis explores the viability and utility of creating small area units that are homogeneous in terms of a specific exposure, neighbourhood active living potential, hypothesised to be associated with a specific health indicator, walking. The thesis has three specific goals: 1) using geographical information systems, develop and test an approach for creating small area units that are homogeneous on a specific exposure, 2) using these new area units, examine associations between neighbourhood active living potential and walking behaviours, and 3) develop individual indicators of exposure to neighbourhood active living potential using an idiosyncratic definition of small areas.

Individual data on walking behaviours and socio-economic characteristics are from a sample of 2923 adults aged 45 years and living in Montreal, Canada. Area data are from the 2001 Canadian census data and from various spatial databases, including Montreal’s land-use (City of Montreal) data, road network data (Société de Trasport de Montréal). Data will be analysed using spatial analysis and multilevel modelling procedures.