Reliability and validity of psychosocial and environmental correlates measures of physical activity and screen-based behaviors among Chinese children in Hong Kong

Table 3 Factor structure and test-retest reliability of the emergent scales in the questionnaire

Scales		% Variance explained	Eigen value	Cronbach's alpha	Test-retest ICC
Self-efficacy		47.0	2.35	0.75	0.82
Variable removed: Walk to school instead of having to wait for a ride
Family and peer support for PA
Factor 1: Family support	Being physically active with whole family; with father; with mother; with sibling	28.2	2.26	0.74	0.86
Factor 2: Peer support	Being physically active with friends; receive encouragement from friends	15.3	1.22	0.50	0.78
Variable removed: Being physical active with grandparents; receive encouragement from family member
Rules and guidance on children's screen-based behaviors
Factor1: Guidance	Guidance during TV, computer games and Internet use	44.0	2.64	0.86	0.83
Factor 2: Rules	Control time in TV, computer & Internet use; no TV during meal time; not allow TV until homework done	20.6	1.23	0.65	0.81
Perceived neighborhood safety		45.5	2.25	0.71	0.89
Social environment in the neighborhood		55.8	2.23	0.73	0.84

PA: physical activity; TV: television
Principal axis factoring with direct oblimin rotation (if two factors are identified). Kaiser-Meyer-Olkin values range from 0.650 (social environment in the neighborhood) to 0.728 (rules and guidance). Bartlett's Test of Sphericity for all the scales are significant at P < 0.01.
Variables that have factor loadings < 0.30 are removed.

ISSN: 1479-5868