Large-scale Assessments in Education (Oct 2022)
Trend analyses of TIMSS 2015 and 2019: school factors related to declining performance in mathematics
Abstract
Abstract Gaining knowledge of what contextual factors may contribute to changes in student achievement across cycles of international large-scale assessments (ILSA), is important for educational policy and practice. Addressing this necessitates advanced methodology that utilizes the trend design of the ILSAs. The Trends in International Mathematics and Science Study (TIMSS) data is suitable for such analyses as it measures students’ competence based on the participating countries curricula. In Norway, students’ performance in mathematics decreased from 2015 to 2019 as evidenced by TIMSS. During this time-period, there are indications that also school climate, student motivation and self-concept decreased. This study investigates whether school climate (including bullying, a safe environment, school emphasis on academic success, and students’ sense of school belonging and well-being), student motivation (including interest-enjoyment and utility value), and self-concept declined from 2015 to 2019, and whether this possible decline is related to the decline in mathematics performance in Norway. The present study utilized a trend approach with mediation structural equation modelling. The results showed that school climate, the utility value of the subject and students’ self-concept declined during this period, and that a safe environment and student self-concept mediated the changes in achievement from TIMSS 2015 to 2019. Hence, declines in a safe school climate and student self-concept were associated with declining achievements. While the study cannot prove a causal relation, it is discussed whether this could indicate that the decline in these predictors may explain the decline in mathematics achievement. The usefulness of the methodological approach for other countries, as well as the implications of the results for policy, practice, and research are discussed.
Keywords