Abstract:

OECD research has shown a problem that there is a relatively small number of students who demonstrate high performance in mathematics, which is a long-term trend in Latvia. Across OECD countries on average in the year 2018 in mathematics 2,4% of students are high-achieving students, meaning that they are proficient at Level 6 (Level 5 and 6 in total 10,9%), whereas in Latvia it was 1,4% (Level 5 and 6 8,5%). In the National Development Plan of Latvia until 2027, one of the targets is to be above the OECD average by increasing the number of high-achieving students.

This research is conducted as part of a larger research project. The aim of the article is to analyze the situation on how national scale assessment tests (large scale assessment tests - diagnostic and exam tests) in mathematics in grades 3, 6, 9 and 12 provide an opportunity to follow the high and low performance of students compared to the PISA test.

The research sample consists of 16 schools from 2 municipalities (in each approximately 350 students per class group) as well as all state students in four grade groups who took these tests in mathematics in 2021. A comparative case study was conducted, the Rasch analysis was used to analyze student performance, 2 independent experts performed performance level pinching using the OECD PISA methodology.

A question was posed: to what extent do the large-scale assessment tests provide an opportunity to see students whose results are high or low and what are the differences between schools in the same municipality?

Overall, data analysis shows a gradual decrease of students with high-performance results and an increase of students with low performance in grades 3., 6., 9., and 12. There are significant differences in the performance of students among schools of the same municipality, already starting from the 3rd grade. At the same time, Rasch's analysis identifies a problem for the proposed assignments, especially in grades 3 and 6, the assignments are considered too easy for the particular study sample, which limits the generalization of the study data.

Keywords:

Mathematics, large scale assessment, performance, high-achieving students.