Table 6. Standardized estimates for the cross-lagged analyses between each smoking-specific parenting strategy and adolescent smoking controlled for age, gender, education, maternal and paternal smoking.

	Frequency of communication 1	Quality of communication 2	House rules 3
Cross-lagged paths
Parenting T1 → Smoking T2	.06	−.19**	−.06
Parenting T2 → Smoking T3	.09	−.04	−.05
Parenting T3 → Smoking T4	.11*	−.11**	.06
Parenting T4 → Smoking T5	−.04	−.14**	−.01
Smoking T2 → Parenting T3	.07	−.07	.01
Smoking T3 → Parenting T4	.07	−.12*	.07
Smoking T4 → Parenting T5	.14*	−.01	.01
Cross-sectional associations
Parenting T2 ↔ Smoking T2	.03	−.15*	.10
Parenting T3 ↔ Smoking T3	.003	−.13**	.01
Parenting T4 ↔Smoking T4	.14*	−.16**	.03
Parenting T5 ↔ Smoking T5	−.06	−.23***	.08
Stability paths
Parenting T1 → Parenting T2	.59***	.56***	.76***
Parenting T2 → Parenting T3	.54***	.61***	.78***
Parenting T3 → Parenting T4	.56***	.62***	.76***
Parenting T4 → Parenting T5	.49***	.69***	.80***
Smoking T2 → Smoking T3	.56***	.55***	.56***
Smoking T3 → Smoking T4	.69***	.68***	.70***
Smoking T4 → Smoking T5	.80***	.75***	.80***

Note.

¹ χ² = (df = 18, N = 365) = 69.50, CFI/TLI = .95/.76, RMSEA = .09;

² χ² = (df = 18, N = 365) = 89.87, CFI/TLI = .94/.74, RMSEA = .11;

³ χ² = (df = 18, N = 365) = 117.17, CFI/TLI = .95/.73, RMSEA = .123;

^***; p<.001,

^** p<.01,

^* p<.05.