Table 4.

Prediction of adolescents’ screen time by single mediation analyses^†

Measurement	n	c-path (SE)	a-path (SE)	b-path (SE)	ab‡	95% CI
TV/DVD (hours/week)
Parental education#	573	0.75 (0.68)
Maternal modelling			3.63 (0.55)**	0.28 (0.05)**	1.01	0.62, 1.65
Paternal modelling			3.25 (0.60)**	0.23 (0.05)**	0.75	0.38, 1.27
Maternal regulation			−0.06 (0.10)	−1.40 (0.29)**	0.08	−0.16, 0.42
Paternal regulation			−0.11 (0.10)	−0.60 (0.28)*	0.06	−0.02, 0.33
PC/game (hours/week)
Parental education#	570	1.36 (0.65)*
Maternal modelling			0.24 (0.36)	0.12 (0.08)	0.03	−0.04, 0.34
Paternal modelling			0.30 (0.41)	0.17 (0.07)*	0.05	−0.07, 0.26
Maternal regulation			0.05 (0.10)	−0.04 (0.27)	0.00	−0.09, 0.05
Paternal regulation			0.05 (0.11)	−0.01 (0.26)	0.00	−0.07, 0.07

SE: Standard Error, CI: Confidence Interval.

Statistical significance at *p>0.05 and **p>0.01.

^† Linear regression analysis, adjusted for adolescents’ gender, weight status and living status at baseline.

^‡ Mediation effects calculated as a-path x b-path.

^# Parental education as high (university/college) vs low at baseline.