PISA 2012 mathematics literacy in Serbia: A multilevel analysis of students and schools

Abstract

In PISA 2012 cycle, the focus was on the mathematics literacy. Data collected in PISA survey in Serbia on 4684 respondents show that students on average have significantly lower performance compared to average OECD performance. The aim of the study was to explore student (variables at the first level) and school level predictors (variables at the second level) of the PISA mathematics literacy using multi-level modelling. The most important finding is that student and school level variables are explaining variability in PISA mathematics performance almost equally. Results show that on the student level, significant predictors are gender, noncognitive characteristics (mathematics anxiety, mathematics self-efficacy, mathematics self-concept, openness for problem-solving), student perceived teaching quality and studying habits. Results also indicate that several school-level variables have direct effects on the PISA math performance and that several school-level variables moderate the relationship between student characteristics and PISA performance.

Keywords

• mathematics literacy
• student-and school-level factors
• multilevel modelling
• pisa 2012