Some insights on the relationship between student performance and test related emotional aspects

Large-scale educational measurement studies usually collect a wide range of data about students’ achievements and collateral information. In this study, we investigate the relation between the mathematics performance of students and background variables, by taking into account the students’ emotional component defined by the test anxiety. INVALSI 2018 data for grades 5, 8, and 10 are considered. Two different approaches are implemented: a parametric one, the latent regression model, and a non-parametric one, namely regression trees. The overall results show that the emotional component plays a key role in explaining the student Mathematics score. In particular, regression trees show that the groups of students associated to the lowest mean performance are also characterized by considerable relevant anxiety levels.