Modifiable neighbourhood units, zone design and residents' perceptions

Area of Science:

• Spatial analysis
• Geographic information systems
• Public health research

Background:

• Neighbourhood effects on health are influenced by how areas are defined, a challenge known as the modifiable areal unit problem.
• Few studies have compared health outcomes across various neighbourhood definitions, including subjective and objective measures.

Purpose of the Study:

• To compare computer-generated neighbourhood zones with those defined by local experts.
• To assess how different neighbourhood definitions align with residents' perceptions and social behaviours.
• To investigate the scale at which neighbourhoods hold meaning for residents.

Main Methods:

• Utilized automated zone design to create synthetic communities based on deprivation and housing type at multiple scales.
• Compared these synthetic zones with subjectively defined community areas in Bristol, UK.
• Matched area definitions against resident perceptions and social behaviour data from the Avon Longitudinal Study of Parents and Children (ALSPAC).

Main Results:

• Automated zone design successfully replicated subjective communities under specific parameter settings.
• Significant, albeit small, differences in neighbourhood perceptions and social behaviour were found across all area sets.
• Resident perceptions aligned equally well with automated zones and subjectively defined communities.

Conclusions:

• Expert-defined neighbourhoods are not necessarily more salient to residents than synthetic areas.
• Local-scale variations in social and housing conditions (at enumeration district level) better explain neighbourhood perceptions.
• Meaningful neighbourhoods for residents are likely smaller than typically studied in geographical health research.