The main objective of this study was to find the role of the student-level and school-level on students’ academic achievement. Research population included students, principals, vice-principals, and academic counselors of Tehran’s vocational and technical schools in the academic year 1391-92. 10 regions from 5 Tehran’s educational districts and 2 schools from each region were selected, with a total of 906 third-grade students from these schools and from 39 classes using multi-stage cluster sampling. 59 school staffs from the same classes and schools were also chosen to take part in this study. Standard AMS and MSLQ questionnaires were used for academic motivation and academic engagement, respectively, as well as researcher-made questionnaires for peer-assisted learning and quality of education. Reliability of the questionnaires were assessed using Cronbachs alpha coefficient, and their content and construct validity was checked using exploratory factor analysis. Academic achievement was calculated using the average of three scores including GPA (grade point average) of the previous year, GPA of three core courses and GPA of practical courses. Data was analyzed using descriptive statistics, correlation matrix, confirmatory factor analysis, structural equation modeling, and multi-level analysis. According to the unconditional model for the school-level, the average academic achievement of students in different schools was significantly different. For the school-level model, 6.13% and 12.05% of the variance of the academic achievement were originated from the student-level and school staff, respectively. In conclusion, 18.81% of the variance for academic achievement was related to the school-level. Findings of the research showed the importance of school level factors -at higher level- influencing other variables in predicting academic achievement.