Ka Man Tam

Year of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Department of of Sport, Physical Education and Health

Principal Supervisor

Cheung Siu Yin.


Social cognitive theory ; Exercise ; Physical fitness ; Electronic measurements




Adults in Hong Kong show relatively low participation in physical activity. In the wake of technological advancements, it has become necessary to promote physical activity in an innovative approach. To that end, this study aimed to investigate the effect of an application of Social Cognitive Theory (SCT) under an eight-week electronic activity monitor system (EAMS)-based intervention on changes in physical activity (PA) as well as its associated SCT constructs of self-efficacy, social support and self-regulation for working adults in Hong Kong. A series of studies were performed: Study 1: In order to assess the validity of the step count output of two popular electronic activity monitor system (EAMS) model, Fitbit Charge HR and Xiaomi Mi Band 2, healthy adult (N=30) worn both EAMS and walked at five predetermined speeds on a treadmill. Two-factor (step x speed) repeated measures ANOVAs was performed to compare the output of devices with manual step count. Result: there was no significant mean difference (p> 0.05) in step count among the Fitbit Charge HR and Mi Band 2 activity monitors and the criterion in all treadmill speeds. Both of them are valid devices for step count in the laboratory setting. Study 2: As to assess the validity of step measurement of Mi Band 2 in the free-living environment, 31 healthy adults were invited for wearing both Mi Band 2 and ActiGraph GT9X Link on their dominant hands wrist for 7 consecutive days. Paired sample t-tests and Pearson correlation were conducted to compare the average steps per day between Mi Band 2 and ActiGraph GT9X Link. Result: there was no significant mean difference (p >0.05) and high positive correlation in step count between the Mi Band 2 and Actigraph. The Mi Band 2 is a valid device for step count in the free-living environment. Study 3: To examine the validity and reliability of the Chinese version of PA related self-efficacy, self-regulation and social support in Hong Kong Chinese adults. There were 230 healthy adults aged 19-63 years recruited. The factorial validity of the scales was assessed by the Confirmatory Factor Analyses (CFA) while criterion validity was assessed by correlating measured constructs with self-reported PA. The internal consistency and scales test-retest reliability were evaluated by Cronbach's alpha and intraclass correlation coefficient, respectively. Result: indicators of CFA supported the one-factor structure while all PA correlates were significant (p<0.01) associated with self-reported PA. All scales demonstrated acceptable internal consistency and test-retest reliability. The results provided psychometric support for using the Chinese version of the scales to measure PA correlates among Hong Kong adults. Main Study: A two-armed randomised controlled trial took place to investigate the effect of an EAMS-SCT intervention on changes in PA and its associated SCT constructs for working adults in Hong Kong. Sixty-four (26 males, 38 females, mean age = 39.98, SD =7.06) participants were stratified based on their job nature and randomly assigned to the control, EAMS only group (n = 33) and the treatment, EAMS-SCT group (n =31) after data screening. Both groups received an EAMS, whereas the EAMS-SCT group received SCT elements in the eight-week intervention, which comprised of one physical activity advisory session conducted by a certified personal trainer and four WhatsApp delivered sessions conducted by the researcher. Outcome measures: Measures of PA (by IPAQ MET score) and SCTs constructs (by the scale of Exercise Self-efficacy, Social Support for Exercise and Physical Activity Self-regulation) of two groups at three times of measurement. 2 x 3 (Group x Time of Measurement) mixed ANOVA and mixed MANOVA were conducted respectively for PA and SCTs constructs for detecting group difference.Results: The interaction effect between different groups (EAMS only and EAMS-SCT) on the participants' score on IPAQ, across three times of measurement (pre-intervention, post-intervention and 4-week follow-up) was significant. The mean IPAQ MET scores of participants in both groups (EAMS only, from 957.64 to 1235.82, EAMS-SCT group from 883.49 to 1420.86) increased and statistically significant difference from its scores than that of the pre-intervention. However, there was no statistically significant difference in IPAQ MET scores between the groups. There was a statistically significant multivariate effect of SCT constructs across the interaction between the groups and time of measurement: F (6, 57) = 7.267, p = .000, with large effect size. Post-hoc analysis revealed that the score of exercise self-efficacy, social support for exercise and physical activity self-regulation from pre- intervention to post-intervention were significantly higher in the EAMS-SCT group. There was no significant difference in the score of physical activity self-regulation between the EAMS only group and the EAMS-SCT group. Meanwhile, there was a significant mean difference in exercise self-efficacy and social support for exercise at Week 8 and social support for exercise at Week 12 between EAMS only and EAMS- SCT group. Conclusion: The participants in the EAMS-SCT group show a significant increase in physical activity level and all three related SCT constructs. Both the EAMS only and the EAMS-SCT groups have an increase in the IPAQ MET score and the score of the scale of physical activity self-regulation after the 8-week intervention. Participants in the EAMS-SCT group exhibit higher scores in the scales of exercise self-efficacy and social support for exercise than participants in the EAMS only group after the 8-week intervention.


Principal supervisor: Professor Cheung Siu Yin ; Thesis submitted to the Department of of Sport, Physical Education and Health


Includes bibliographical references (pages 179-215)