Discrete time multilevel logistic hazard regression for transition to established from experimental smoking using multiple imputation for missing teacher data
| Fixed effects | Estimate | se | z | p Value | HR (95% CI) |
| Intercept | −3.47 | 0.282 | −12.311 | 0.000 | 0.03 (0.01 to 0.06) |
| Wave 2–wave 3 | −0.01 | 0.157 | −0.047 | 0.962 | 0.99 (0.73 to 1.35) |
| Wave 3–wave 4 | 0.46 | 0.158 | 2.934 | 0.003 | 1.59 (1.16 to 2.17) |
| IG-no participation and control group | 0.37 | 0.187 | 1.986 | 0.047 | 1.45 (1.00 to 2.10) |
| Age | −0.02 | 0.091 | −0.258 | 0.797 | 0.98 (0.81 to 1.17) |
| Female | −0.12 | 0.139 | −0.861 | 0.389 | 0.89 (0.67 to 1.17) |
| Sensation-seeking | 0.65 | 0.184 | 3.546 | 0.000 | 1.92 (1.34 to 2.76) |
| Parenting | −0.11 | 0.130 | −0.868 | 0.385 | 0.89 (0.69 to 1.16) |
| Parent smoking | 0.11 | 0.150 | 0.729 | 0.466 | 1.12 (0.83 to 1.50) |
| Peer smoking | 0.86 | 0.126 | 6.863 | 0.000 | 2.37 (1.85 to 3.03) |
| Sibling smoking | 0.19 | 0.137 | 1.408 | 0.159 | 1.21 (0.92 to 1.59) |
| Student-rated class climate | 0.10 | 0.152 | 0.689 | 0.491 | 1.11 (0.82 to 1.50) |
| Student smoking class level | 0.40 | 0.166 | 2.439 | 0.015 | 1.50 (1.08 to 2.08) |
| Teacher-rated class climate | 0.11 | 0.252 | 0.451 | 0.652 | 1.12 (0.68 to 1.84) |
| School type | −0.25 | 0.210 | −1.183 | 0.237 | 0.78 (0.51 to 1.18) |
| Sample sizes | Wave 1 to wave 2 | Wave 2 to wave 3 | Wave 3 to wave 4 |
| Students | 1226 | 1095 | 796 |
| Classrooms | 201 | 178 | 164 |
| Schools | 83 | 81 | 77 |
IG, intervention group.