Table 3

Discrete time multilevel logistic hazard regression for transition to established from experimental smoking using multiple imputation for missing teacher data

Fixed effectsEstimatesezp ValueHR (95% CI)
Intercept−3.470.282−12.3110.0000.03 (0.01 to 0.06)
Wave 2–wave 3−0.010.157−0.0470.9620.99 (0.73 to 1.35)
Wave 3–wave 40.460.1582.9340.0031.59 (1.16 to 2.17)
IG-no participation and control group0.370.1871.9860.0471.45 (1.00 to 2.10)
Age−0.020.091−0.2580.7970.98 (0.81 to 1.17)
Female−0.120.139−0.8610.3890.89 (0.67 to 1.17)
Sensation-seeking0.650.1843.5460.0001.92 (1.34 to 2.76)
Parenting−0.110.130−0.8680.3850.89 (0.69 to 1.16)
Parent smoking0.110.1500.7290.4661.12 (0.83 to 1.50)
Peer smoking0.860.1266.8630.0002.37 (1.85 to 3.03)
Sibling smoking0.190.1371.4080.1591.21 (0.92 to 1.59)
Student-rated class climate0.100.1520.6890.4911.11 (0.82 to 1.50)
Student smoking class level0.400.1662.4390.0151.50 (1.08 to 2.08)
Teacher-rated class climate0.110.2520.4510.6521.12 (0.68 to 1.84)
School type−0.250.210−1.1830.2370.78 (0.51 to 1.18)
Sample sizesWave 1 to wave 2Wave 2 to wave 3Wave 3 to wave 4
 Students12261095796
 Classrooms201178164
 Schools838177
  • IG, intervention group.