This study focuses on the research area of scholastic efficacy. It postulates the existence of certain characteristics of schools which have a personal and significant effect on the scholastic achievement of their students. In particular, the work falls into the category of studies on school effects. The effect of the school is defined as the proportion of the variance of the scholastic achievement results attributable to school policies and practices. The purpose of the study is to estimate the magnitude of the school’s effect, i.e. the relative contribution of the school in the total variation of academic achievement in Spanish and Mathematics in the three grades of the Mexican junior high school. The data employed come from the National Standard Tests for 2002 and 2003, applied to a national sample of junior high schools in Mexico. The analysis was performed with the HLM program, using Multilevel Models. The work was done with a model of two levels (student and school). The research, which is descriptive and correlational, contributes explanatory elements based on the analyses made. The key findings are these: a) The estimated effect of the school through the intraclass correlation index (ICI) was 12% for Spanish and 9% for Mathematics, which means that the Mexican junior high schools are quite similar to each other. Although it may seem paradoxical, we have confirmed that this effect, numerically speaking, is sufficient for us to say that the school which students attend makes the difference. b) In comparing scholastic effects between the junior high schools’ modalities of operation, it was found that in the case of Mathematics, the telesecundarias (junior high schools conducted by Internet) have a greater effect on student performance than do general and technical public schools, or private schools, in both Spanish and Mathematics.

Key words: School effectiveness, secondary education, mathematics instruction, language Instruction.