Background
Physical activity is one of the key factors of obesity prevention in children and adolescents. To encourage children and adolescent to participate in moderate to vigorous physical activity (MVPA) is a priority of public health practice. However, the PA of children and adolescents are reported to decline in early adolescence. Although studied have found that PA of adolescents may be influenced by personal physical ability, self-perceptions and environmental factors, what factors predict objectively measured MVPA in adolescents remains unclear. Therefore, the purpose of this study was to examine physical ability (body mass index [BMI] and physical fitness [PF]), self-perceptions (enjoyment of PA [EPA], attitudes towards physical education [ATPE], and self-competence in sports [SPCS]), social (school teacher encouragement, peer encourage and involvement, and family facilitation and role modeling) and environmental (school sports facilities, neighborhood sports facilities and parks) factors that influence adolescents’ participation in MVPA in Brunei Darussalam. Bandura’s social cognitive theory was used to be the framework for examining the determinants of PA behavior since it is based on reciprocal, dynamic and triadic relationships between personal, social and environmental factors and PA behaviors.

Objectives
To examine the personal, social and environmental predictors of MVPA in Bruneian adolescents.

Methods
The study was approved by University Ethics Committee and Ministry of Education of Brunei Darussalam.  A total of 468 12-14 year old Bruneian adolescents (boys = 200; girls = 268) from eight Brunei secondary schools voluntarily participate in this study. Subjects’ body height and weight were collected. Each subject wore the RT3 accelerometer for seven consecutive days to objectively measure light PA (LPA), moderate PA (MPA), vigorous PA (VPA) and MVPA. PF including six fitness components (push-up, sit-ups, sit and reach, standing broad jumps, shuttle run, and 1.6km run) were measured. BMI was calculated using body weight ∙height–2. A Physical Activity Questionnaire was administrated to examined personal, social and environmental factors. The questionnaire consisted 25 items that investigated the adolescents’ PA patterns and lifestyle, school sports facilities, neighborhood facilities and parks, teachers’ and peers’ encouragement and involvement and parent’s influences. Face and content validity of the instrument was established by a panel of three experts. One-week test-retest reliability of the instrument was established (r = .81) and internal consistency for this instrument was demonstrated (Cronbach’s Alpha [α] = .75). The Physical Activity Enjoyment Scale (PACES) (α = .87), ATPE (α = .69) and SPCS (α = .86) were administrated. Social Economic Status (SES) was provided by school administration.

Results
All the adolescents in the study did not meet recommended PA guideline (MVPA: 4.07 ± 3.45 minutes∙day–1) and girls were more inactive than boys (girls MVPA: 2.93 ± 2.26 minutes∙day–1; boys MVPA: 5.60 ± 4.13 minutes∙day–1, t = 8.95, p < .001). BMI did not significantly predict all components of PA while PF significantly predicted high intensity of MVPA. Results from correlation analysis showed that PF was significantly related to MVPA (r = .19, p = .001), the ATPE (r= .23, p = .001) and PACES scores (r = .21, p = .001) while BMI was only related to the PACES scores (r= -.18, p = .001). Results from multiple stepwise regression analysis revealed that family doing PA together, doing fitness enhanced exercise, school sports facilities, proximity of neighborhood sports facilities and parks, friends doing PA together, total ATPE attitude scores and PF significantly predicted MVPA. These factors explained a significant proportion of variance in the PACE scores (R2 = .17, F(1, 462) = 5.11, p = .02, SEE = 3.16) while family doing PA together, school sports facilities and proximity of sports facilities and parks, watching TV, watching DVD and playing computer games, friends doing PA together and total enjoyment levels significantly predicted LPA (R2 = .25, F[1, 454] = 6.06, p = .01, SEE = 19.43).

Conclusions
PF but BMI was a significant predictor of adolescents’ participation in MVPA. Attitude but enjoyment was a significant predictor of MVPA while enjoyment was a significant predictor of LPA. Youth individual fitness levels, days of doing fitness enhanced exercise, parent role modeling, peer involvement, school sports facilities, neighborhood facilities and parks all play important roles in the youth’s participation in MVPA.

Implications
Health educators are suggested to encourage youth to improve and maintain physical fitness through participating in more MVPA. The findings from this research highlighted that personal (PF and attitude), social (parental role modeling) and environmental (school sports facilities and neighborhood facilities and park) factors were influencers of youths’ participation in MVPA. This study suggests government policy makers to improve sports environments in both schools and communities. It also appeals to adolescents and their parents to be aware of their PF levels, parental participation and role modeling in encouraging their children to participate in MVPA.