Faculty of Science

Study of Environmental and Individual Determinants (SEID)

Further Information

Research team

Chief Investigators
Associate Investigator
Project Coordinator

Audit tool

Files for download

SPACES Audit Manual [PDF, 3.2 MB]
Updated 27 Aug 2009

The investigation of the relationship between the physical environment and physical activity is relatively new and evidence collected to date has been limited.

The Health Promotion Research and Evaluation Group began a study in 1999, funded by National Health and Medical Research Council (NHMRC), to examine the relationship between the physical environment and walking for recreation and for transport (known as SEID II). This study followed SEID I which examined perceptions of the environment, and the individual, social and physical environmental factors associated with recreational physical activity and walking in 1773 healthy, sedentary working adults and home makers.

Based on the results of in-depth interviews and a Delphi process, conceptual frameworks were developed to examine the physical environmental factors relevant for walking for recreation and for transport. These factors were arranged into three levels: features (overall factors that influence the physical environment), elements (factors that are components of the features) and items (factors that have the potential to be changed to improve an element).

Four features were identified as influencing walking behaviour: functional (for example, sidewalks), safety, aesthetics, and destinations. Using this framework, an observational instrument was developed to measure the environment in local neighbourhoods and was used to collect data over a 408 square kilometre area within the metropolitan area of Perth, Western Australia. Additional environmental information was obtained from other sources. The data were aggregated to provide neighbourhood scores for each of 1773 randomly-selected respondents from SEID I who reported their physical activity. The items, elements and features were weighted according to average weights provided by the Delphi participants.

Only feature level data are presented. Using logistic regression, the environmental features were examined for correlates for walking near home.

Walking for recreation

After adjusting for demographic, individual and social factors, functional features (Odd Ratio (OR) 1.62; 95% Confidence Interval (CI)1.20-2.19) appeared to be more important for walking for recreation near home than safety (OR 1.20; 95% CI 0.92-1.57), aesthetic (OR 1.20; 95% CI 0.84-1.73) or destination (OR 0.99; 95% CI 0.71-1.36) features.

These functional items include the presence of a well-maintained, continuous sidewalk system; the design of the street system; and traffic volume, speed and traffic calming devices.

Walking for transport

A different set of factors were related to walking for transport near home. After adjustment, destination features (OR 1.80; 95% CI 1.33-2.44) appeared to be more important for walking for transport near home than functional (OR 1.30; 95% CI 0.97-1.73), safety (OR 1.12; 95% CI 0.85-1.46) or aesthetic (OR 0.90; 95% CI 0.66-1.22) features.


The findings suggest those neighbourhoods with attractive and comfortable pedestrian facilities - safe, convenient places to walk; narrow streets, slower traffic, and verges that separate motor vehicles from pedestrians, and local destinations - were associated with walking.

The environmental factors identified in this study will assist both policy makers and practitioners across several sectors in the creation of neighbourhood environments that are conducive for walking.

Back to top