Table 2.

Description and psychometric properties of hypothesized mediator scales (Australia, 2011)

	Description of scale	Range (no. of items)	Source	α
Barrier self-efficacy	* Single factor 5-point Likert format	1-5	Adapted version of an 8-item questionnaire previously developed for use with 5th, 8th and 9th grade girls (PASES) [46]–[48]. The modified scale has been shown to be a valid measure of barrier self-efficacy for this age group (α=.81, ICC=.57) [49], and confirmatory factor analysis showed good fit for use with 6th and 8th grade girls (CFI=.98; CFI=0.99) [47]. Factor structure, loadings, factor variance, item means and errors were shown to be invariant across age groups and race / ethnic groups (SE=0.4, 0.024, p<0.001) [50,51], with good test-retest stability (.84) [49].	1= .75
	* Participants were asked to select how much they agree with the eight statements by ticking the relevant circle	8 items
	* E.g. I can be physically active even if it is hot or cold outside).
	* Scale: 1=Disagree a lot to 5= Agree a lot
Enjoyment	* 5-point Likert format The child was asked to select how often they experience the relevant feeling about physical activity by ticking the relevant circle	1-5	Adapted version of the a 16-item version of the Physical Activity Enjoyment Scale (PACES) [52] and has been recently validated for use with children (CFI=0.95), with good stability across age groups (SE=0.03;0.24, p<0.001) [52]–[54] and good test-retest stability (.73) [49].	1= .72
		6 items
	* E.g. When I am physically active…….…. It’s no fun at all	(negatively worded)
	* Scale: 1=Never to 5= Every day
Social Support family peers/friends teacher	* 5-point Likert format	1-5	Adapted scale based on two scales used in the student survey of the Amherst Health and Activity Study [55]. Recently tested for validity and use with children in the 6th and 8th grade by Dishman and colleagues (family and friend scales only) [51]. Validity measures indicate that the factor structure, factor loadings and factor variances / co-variances were invariant across race/ethnic groups and across age groups and across time (CFI=0.96; 0.98, SE (friends) =0.41; .027 and SE (family) =0.53; 0.021, p<0.001). The teacher social support scale was devised for the purpose of this study and follows the structure and wording of the family and friends social support for physical activity scales [51].	P1= .68
	* Participants were asked to select how often a specific form of social support (encouragement or modeling) is provided to them during a typical week by ticking the relevant circle	3 scales
				F1= .65
		Peers (P)
		(3 items)
		Family (F)		T1= .77
		(4 items)
		Teacher (T)
		(4 items)
	* E.g. Modeling: During a typical week at school, how often do your FRIENDS.... do physical activity or play sports with you?
	* E.g. Encouragement: During a typical week at school, how often does your TEACHER.... encourage you to do physical activity during recess or lunch breaks?
	* Scale: Never = 1 to Every day = 5
Perceived school enviornment	* Single factor 4-point Likert format	1-4	Adapted version of the 2-factor, 20-item questionnaire Q-SPACE developed by Robertson-Wilson, Levesque and Holden [56]. Initial findings support the reliability (α=0.86 and test-retest reliability=0.78) and construct validity of the physical environment sub-scale for use with children and youth [56,57].	1= .80
	* The participant was asked to select how much they agree with the eight statements by ticking the relevant circle	9 items
	* E.g. There is sports equipment available for students to use during recess and lunch breaks
	* Scale: 1= Strongly Disagree to 4= Strongly Agree (no neutral response)