Math anxiety-contributing school and individual level factors

Abstract

PISA 2003 survey data indicate high levels of mathematics anxiety among students in Serbia. More than a half of Serbian students are concerned with whether they will have difficulties in a mathematics class or earn poor marks. At the same time, the achievement on the mathematical literacy scale is very poor. Building on control-value theory, the analysis in this paper focuses on exploring possible correlates of math anxiety, separating the school level factors from individual level factors (HLM), and differences between different groups of students in respect to their relationship towards mathematics (hierarchical cluster analysis). Data suggest high levels of anxiety within the Serbian student body to be a rather systemic problem not contributed to a particular school. Approximately 6 % of the total variance in math anxiety is explained by the between schools differences, while 94 % of differences accounts for the within the school variance. The achievement and interest in mathematics, high mathematics self-concept, and school and classroom atmosphere are associated with a lower level of math anxiety. The only school level factor that has a minor, but significant effect is the index of economic, social, and cultural status. Based on students' attitudes towards mathematics, they can be divided into three clusters. Dimensions that distinguish the clusters are interest in mathematics and presence of math anxiety. The group displaying anxiety, scores the lowest among the three in math achievement.