Investigation of student and teacher's characteristics to predict students' mathematics anxiety and achievement with hierarchical linear modeling

In this study, it was aimed to determine the predictive power of student and teacher characteristics in students' math anxiety and achievement. For this purpose, the variables of students and teachers, as well as student variables related to students' math anxiety and achievement, were handled at two levels using hierarchical linear modeling. In this respect, a predictive relational research design based on the analysis of direct and indirect effects was used in the research. The research sample consists of mathematics teachers (12), students of these teachers (413) and parents of these students (413) working in Osmaniye province in 2022-2023. Mathematics scorecard achievement scores at the end of the semester were used for students' success level, "Mathematics Anxiety Scale" for students, "Mathematics Anxiety Scale for Teachers" and "Parent Mathematics Anxiety Scale" were used for anxiety levels. As a result of the two-level analysis, the variables related to student mathematics achievement; parental anxiety, student gender, parent education level, student mathematics self-efficacy perception and teacher level variables are teacher mathematics anxiety, teacher gender, teacher marital status. Variables of student level associated with student math anxiety; parental anxiety, student's social opportunity, student mathematics self-efficacy perception and teacher level variables are teacher mathematics anxiety, teacher gender, teacher marital status, teacher education level. It can be stated that the majority of the variability in students' mathematics achievement and anxiety depends on the student level variables, while the remaining part of the variables can be explained by the teacher level.