General descriptive statistics will be used to describe the
baseline characteristics of the participants. As the adolescents
are nested in schools we will use a multilevel regression
approach with three levels to assess the effects
of the intervention on behavior. The first level are the
repeated measures within the participants (baseline and
two follow-up measurements), the second level are the
pupils and the third level are the schools, where the pupils
are nested. Multilevel linear regression will be used
to analyze the effects of the program on week consumption
and intention and multilevel logistic regression will
be used to analyze the effects on binge drinking and excessive
drinking. As covariates we will include condition,
age, gender, educational background, religion, and the
outcome variable on baseline. Moderation analyses will
be performed to assess different effects for low and high
educational level, age, and gender.