Identifying factors influential on 8th grade Iranian students mathematics achievement in TIMSS 2011

Using the data of TIMSS 2011 and the statistical methods, namely hierarchical linear modeling, in this study we suggested a mathematic achievement model for Iranian students in terms of the selected variables. The data was collected from 4837 students nested in 227 schools. Variables were entered into the model at two levels, i.e., student (gender, parents’ education level, number of books available at home, student’s perception of school, mathematic values, selfconcept of ability in mathematics) and teacher/school (teacher’s experience, teacher’s education level, teacher’s perception of school, teacher’s cooperation with others) to demonstrate the most important mathematical achievement predictors. The analysis of the data using the hierarchical linear model showed that among the levels of the student variable, the self-concept of ability in mathematics, numbers of books available at home, and student’s perception of school; and among the levels of the teacher/school variable, teacher’s experience and his perception of school predicted students’ mathematics achievement. Finally, the Mathematical Progress Model indicated the most important variable as “self-concept of ability in mathematics” and it showed that 1 unit increase in self-concept of ability in mathematics leads in 46.05 unit increase in the mathematics achievement level. This model also decreased the18.2 percent of “within-school” variances and 16.3 percent of “between-school” variances, in comparison with the first model.