Date of Award


Degree Type


Degree Name

Doctor of Philosophy


School of Exercise, Biomedical and Health Sciences


Faculty of Computing, Health and Science

First Advisor

Dr Elizabeth Rose

Second Advisor

Dr Kay Cox


Australia has an increasingly aging population with increasing levels of physical inactivity. The potential detrimental effects of these two factors on the health of the community highlight the need to investigate methods to increase physical activity in older Australian adults. The study reported in this thesis formed part of the PATH (Physical Activity Time for Health) Project, a community-based research trial that compared two strategies to increase physical activity in underactive, 60-80 year old men (n = 66) and women (n = 188). Twelve recreation centres were randomised to either a supervised group based walking intervention with behavioural change components, or a self-managed/usual care intervention. Participants in behavioural intervention centres were asked to complete 150mins/week of moderate intensity physical activity as a supervised walking program, organised as 3 sessions/week for the first 3-months and then 1 supervised and 2 unsupervised walk sessions/week for the second 3-months. Participants in self-managed centres were asked to complete 3 sessions of moderate intensity physical activity (150mins/week) for 6 months. In this thesis I have investigated the efficacy of Self-Determination Theory (SDT; Deci, 1980) to explain motivation of older adults to adopt physical activity. There were three sub-purposes. First, to determine the effect of the behavioural intervention compared with the self-managed approach on psychosocial, physiological, and physical activity outcomes. Second, to investigate the contributions of psychosocial predictors to adherence and physical activity level across the self-managed and behavioural intervention strategies. Third, to estimate the directional relationships between self-determination constructs and adherence using path analysis. The physical activity outcomes measured in this study were retention, adherence and total physical activity level. Retention was defined as the number of participants in the study after 6 months. Adherence was defined as the number of exercise sessions completed over the 6 months. Total physical activity level was measured using the Physical Activity Scale for the Elderly (PASE; Washburn, Smith Jette, & Janney, 1993). At baseline the behavioural intervention program had 138 participants, compared to 116 participants in the self-managed condition. After 6 months the behavioural intervention program had retained 84% of these participants, compared to 67% in the self-managed condition. With respect to adherence there was no significant difference between participants in the behavioural intervention compared to those in the self-managed condition (67.7% and 59% of sessions, respectively). The total physical activity level (related closely to the adherence score) also did not differ between conditions (114.69 and 115.87 for the behavioural intervention and self-managed groups, respectively). The major and novel finding of this study was that social connectedness was a significant factor in the engagement of older adults in physical activity. This was evidenced by the increases in social connectedness in the behavioural intervention group, compared to decreases in social connectedness reported in the self-managed group. Furthermore, structural equation modelling demonstrated that social connectedness, compared to physical self-perceptions and autonomy, was the only significant predictor of adherence. This study also found that self-perceptions outside the physical domain can have as important a role in exercise behaviour as physical self-perceptions. Structural equation modelling provided further support for this proposition showing adherence was more strongly related to social self-perceptions than physical or cognitive selfperceptions after the intervention. Also, lower perceptions of physical appearance and higher perceptions of nurturance were associated with higher total physical activity levels at 6-months This study confirms previous research and contributes novel findings demonstrating the importance of social connectedness in physical activity behaviour in older adults. Further it provides strong evidence for the ability of physical activity to influence multiple aspects of the lives of older adults. These findings have implications for health practitioners and development of policy and programs to increase physical activity. Employing Self-Determination Theory has further elucidated motivation for exercise in older adults and provided novel findings to support inclusion of socially based components into physical activity promotion campaigns for older adults.