The relationship between high-stakes assessments and anxiety: Consequences for academic performance

The use of high-stakes assessments has become widespread in educational systems. While they can potentially boost student achievement, they may also increase anxiety levels, which could negatively affect their academic performance. These effects might not be uniform, varying across schools and among distinct student profiles. In this paper we examine the relationship between anxiety and performance and investigate how high-stakes assessments influence students' anxiety levels and academic outcomes. We conduct our analysis at the average level and by several individual characteristics, estimating a series of models using data from an international sample of countries. This allows us to shed new light on the winners and losers of the policy, and to get a deeper understanding of the role of anxiety. The study provides first international evidence by exploiting novel cross-country information on high-stakes testing. Our results extend previous research by revealing a strong negative correlation between student anxiety and academic performance, regardless of whether the country implements high-stakes testing. Furthermore, the heterogeneity analysis by socioeconomic level indicates that students from more advantaged socioeconomic backgrounds exhibit a stronger negative association between anxiety and performance.