An objective index of walkability for research and planning in the Sydney Metropolitan Region of New South Wales, Australia: an ecological study
1 Sydney School of Public Health, The University of Sydney, Sydney, NSW 2006, Australia
2 Public Health Unit, Illawarra Shoalhaven Local Health District, Wollongong, NSW 2522, Australia
3 Illawarra Health and Medical Research Institute, University of Wollongong, Wollongong, NSW 2522, Australia
4 University Centre for Rural Health - North Coast, Sydney School of Public Health, The University of Sydney, Sydney, NSW 2006, Australia
5 North Coast Public Health Unit, Lismore, NSW 2480, Australia
6 Bureau of Transport Statistics, Transport for NSW, Haymarket, NSW 1240, Australia
7 New South Wales Ministry of Health, North Sydney, NSW 2060, Australia
8 Centre for Research, Evidence Management and Surveillance, Sydney and South Western Sydney Local Health Districts, Liverpool, BC NSW 1871, Australia
9 School of Public Health and Community Medicine, University of New South Wales, Liverpool, BC NSW 1871, Australia
10 Centre for Health Research, University of Western Sydney, Campbelltown, NSW 2560, Australia
11 National Centre for Epidemiology and Population Health, The Australian National University, Canberra, ACT 0200, Australia
International Journal of Health Geographics 2013, 12:61 doi:10.1186/1476-072X-12-61Published: 24 December 2013
Walkability describes the capacity of the built environment to support walking for various purposes. This paper describes the construction and validation of two objective walkability indexes for Sydney, Australia.
Walkability indexes using residential density, intersection density, land use mix, with and without retail floor area ratio were calculated for 5,858 Sydney Census Collection Districts in a geographical information system. Associations between variables were evaluated using Spearman’s rho (ρ). Internal consistency and factor structure of indexes were estimated with Cronbach’s alpha and principal components analysis; convergent and predictive validity were measured using weighted kappa (κw) and by comparison with reported walking to work at the 2006 Australian Census using logistic regression. Spatial variation in walkability was assessed using choropleth maps and Moran’s I.
A three-attribute abridged Sydney Walkability Index comprising residential density, intersection density and land use mix was constructed for all Sydney as retail floor area was only available for 5.3% of Census Collection Districts. A four-attribute full index including retail floor area ratio was calculated for 263 Census Collection Districts in the Sydney Central Business District. Abridged and full walkability index scores for these 263 areas were strongly correlated (ρ=0.93) and there was good agreement between walkability quartiles (κw=0.73). Internal consistency ranged from 0.60 to 0.71, and all index variables loaded highly on a single factor. The percentage of employed persons who walked to work increased with increasing walkability: 3.0% in low income-low walkability areas versus 7.9% in low income-high walkability areas; and 2.1% in high income-low walkability areas versus 11% in high income-high walkability areas. The adjusted odds of walking to work were 1.05 (0.96–1.15), 1.58 (1.45–1.71) and 3.02 (2.76–3.30) times higher in medium, high and very high compared to low walkability areas. Associations were similar for full and abridged indexes.
The abridged Sydney Walkability Index has predictive validity for utilitarian walking, will inform urban planning in Sydney, and will be used as an objective measure of neighbourhood walkability in a large population cohort. Abridged walkability indexes may be useful in settings where retail floor area data are unavailable.