Description: This paper reports on a study about the impact of the school on the development of academic self-concept and language achievement in secondary school. Three research questions will be answered: 1. How large are the school differences with regard to language achievement and academic self-concept ? We will look at the school level variance in the intercept (outcome at the start of secondary education) and the slope (development between Grade 7 and Grade 12) of these two student outcomes. 2. How large is the impact of the school on the relation between academic self-concept and achievement? Building on the reciprocal effects model (Marsh & Yeung, 1997; Van Damme et al., 2004), a positive correlation between academic self-concept and achievement is assumed. Despite the numerous studies on this reciprocal effects model, the effect of the school has been seldom studied. 3. Is there evidence for differential school effectiveness with regard to student gender? In other words, are the school effects on language achievement, academic self-concept and their correlation different for boys and girls? Methodology: Data on language achievement and academic self-concept were gathered through repeated assessement of 2826 students in 50 secondary schools. These adolescents (2196 girls and 1649 boys) filled out a questionaire regarding their academic self-concept in Grade 7, Grade 8, Grade 10 and Grade 12. Their achievement in Dutch (most students' mother tongue) was assessed five times between Grade 7 and Grade 12 (De Fraine et al., 2003). These longitudinal data are analysed by means of a multilevel multivariate growth curve model. The multilevel aspect of this model makes it possible to estimate the importance of the school (school level variances). In order to investigate the differences between boys and girls, a multi-group model was implemented. Conclusions: It was found that schools have a larger impact on students' achievement than on their academic self-concept. About 54% of the variance in the Dutch intercept was situated at the school level, whereas schools account for only about 8% of the variance in academic self-concept. This was also the case for the school effect on the development of achievement and academic self-concept. The correlation between academic self-concept and achievement was larger at the school level than at the student level. The positive correlation at the school level is an indication of consistency of school effects across effectiveness indicators. Schools that are effective for language achievement, tend to be also effective for their students' academic self-concept. Several indications of different school effects for boys and girls were found. The school effects on academic self-concept were larger for girls than for boys, whereas the school effects on achievement were larger for boys. Moreover, the correlation between the two outcomes at school level (the consistency) was larger for boys than for girls.