INTRODUCTION: The aim was to examine the prevalence of mental health difficulties and prejudices toward mental illness among adolescents, and to analyze possible school and school class effects on these issues. METHODS: The sample comprised 4,046 pupils (16-19 years) in 257 school classes from 45 Norwegian upper secondary schools. The estimated response rate among the pupils was about 96%. Self-reported mental health difficulties were measured with a four-item scale that covered emotional and behavioral difficulties. Prejudiced attitudes toward mental illness were assessed using a nine-item scale. Multilevel regression analysis was used to estimate the contribution of factors at the individual level, and at the school and class levels. RESULTS: Most of the variance in self-reported mental health difficulties and prejudices was accounted for by individual level factors (92-94%). However, there were statistically significant school and class level effects (P < 0.01), confounded by socioeconomic factors. Mental health difficulties were commonly reported, more often by females than males (P < 0.01). Difficulties with emotions and attention were the two main problem areas, with definite to severe difficulties being reported by 19 and 21% of the females, and by 9 and 16% of the males, respectively. Prejudices were reported more often by males than females (P < 0.01). Both self-reported mental health difficulties and prejudiced attitudes were related to educational program, living situation, and parental education (P < 0.01). CONCLUSION: The relatively high prevalences of mental health difficulties and prejudiced attitudes toward mental illness among adolescents indicate a need for effective mental health intervention programs. Targeted intervention strategies should be considered when there is evidence of a high number of risk factors in schools and school classes. Furthermore, the gender differences found in self-reported mental health difficulties and prejudices suggest a need for gender-differentiated programs.