Degree Name

Doctor of Philosophy


University of Wollongong Dept. of Psychology


Physical inactivity is associated with increased risk of developing chronic diseases. evidence linking physical activity and health outcomes has led to the development of national physical activity guidelines, which focus on moderate-intensity activities.Walking is the most commonly reported moderate-intensity activity and is arguably the most relevant activity for public health interventions targeting adults.

Public health strategies to increase participation in physical activity are now starting focus on supportive factors in the physical environment. A behaviour-specific ecological approach was taken in this thesis. Ecological models of physical activity behaviour identify multiple levels of influence from intrapersonal, interpersonal, social and broader environmental domains, with particular attention to the physical environment.

A literature review of studies that assessed associations of environmental attributes physical activity found research on environmental influences shows promise to identify significant and potentially modifiable influences on physical activity in general, and specific to walking behaviour. Studies were found to be limited to cross-sectional design and a need was identified for both prospective and intervention studies in order to advance the field.

The main aims of this thesis were to develop and test measures of perceptions of environmental attributes; to examine cross-sectional relationships of perceptions of environmental attributes and an objective measure of location with walking behaviour and, to examine prospective relationships of changes in environmental perceptions with changes in walking behaviour. The studies of these relationships are reported for a workplace sample in Part 3 and for a community sample in Part 4.

The cross-sectional study of Part 3 found for men, significant positive relationships 'aesthetics' 'convenience' 'access' to services, and for coastal place of residence with neighbourhood walking; and a negative relationship emerged for 'traffic' as a problem with neighbourhood walking. For women, a significant positive relationship was reported for 'convenience' and a negative relationship with 'access' to services with neighbourhood walking. Fewer significant relationships were found for total walking and total physical activity.

The prospective study of Part 3 found for men, improved perceptions of 'convenience' and 'aesthetics' were positively associated with increased neighbourhood walking, and improved perceptions of 'access' to services were negatively associated with increased walking. For women, improved perceptions of 'convenience' and 'traffic' were positively associated with increased neighbourhood walking.

Part 4 expanded on the studies of Part 3 by increasing the range of perceptions of the environment measured, and by examining a greater number differentiated walking outcome measures including: neighbourhood walking, walking for exercise, pleasure and walking to get to and from places. A principal components analysis found four main factors underlying the items measuring environmental attributes, these were: 'aesthetics' 'accessibility' 'safety' and 'weather'. The cross-sectional study of Part 4, found more significant relationships of environmental perceptions with walking for men compared to women. In this study, living in a coastal location was associated with w o m e n being more likely to do more neighbourhood walking. Different environmental attribute categories were found related to different types of walking. 'Weather', 'aesthetics', 'accessibility' and 'location' were associated with neighbourhood walking. 'Weather' and 'aesthetics' were found to be associated with Walking for exercise. 'Safety' and 'accessibility' were associated with walking for pleasure. None of the environmental attributes were found to be associated with walking to get to and from place to places.

Results for the prospective study of Part 4 found baseline perceptions of environmental attributes were related to perceptions at follow-up, but not at a multivariate level of analysis. Changes in perceptions of the neighbourhood environment over time were not found to be associated with changes in any of the four walking outcomes. The variation in findings of the two prospective studies may be due to differences between the samples in age, educational attainment, methods of assessment and sample size.

The diverse directions of association emphasize that further work is needed on the development and refining of measures of perceptions of the environment. Future research should also compare measures of perceptions with objectively measured environmental attributes to provide evidence of validity for self-report perceptions. strong gender differences found in the studies of this thesis highlight the need to carry out gender-specific analyses in physical activity studies.

This thesis has overall, given support to the importance of examining the influence of environmental factors on adults' walking behaviour. It is recommended that future research now needs to go beyond looking at environmental variables on their own, and include the strongest individual and social factors in future multi-level prospective studies. Even if small amounts of variance in physical activity are explained by environmental factors, these many small effects across communities could accumulate to mean substantial physical activity changes across populations.

02Whole.pdf (3693 kB)