Abstract
BACKGROUND: Psychological attention to the role of academic performance in university admission tests is still growing. This attention includes both the psychometric characteristics of these tests and the cognitive and non-cognitive factors that may affect performance in admission tests. OBJECTIVE: This study aims to assess the stability and changes in the academic performance of students across two consecutive admission tests. It also aims to examine how participants' genders and languages can relate to student academic performance. DESIGN: To test the research hypotheses, we sampled a large group of school graduates (n = 25.563) who applied and took two national university admission tests in 2024 in Kazakhstan. RESULTS: The results of the Pearson correlation revealed that there is a positive association between the academic performance of applicants in two consecutive national admission test scores (r = .913). The paired t-test results suggested that they were statistically signiicant, with the largest increase in overall scores (AM = .96). Gender was found to affect the academic performance such that females outperformed males (85.6% vs. 64.6% pass rates), showing stronger effect sizes. There were also statistical differences between the two groups of Russian and Kazakh test takers; however, this difference was not practically significant. CONCLUSION: These findings suggest that demographic characteristics of applicants can considerably influence the university admission results, and it is recommended that test developers be attentive to this issue. Academic performance showed strong stability over time, especially in STEM subjects, with significant but modest overall improvements across subjects.