Assessment in STEM Education: A Bibliometric and Systematic Review of Methods, Innovations, and Interdisciplinary Approaches

Abstract

STEM (Science, Technology, Engineering, and Mathematics) education plays an essential role in preparing students for technological advancement. However, selecting appropriate assessment methods remains challenging across educational levels. This study employed a systematic literature review (SLR) and bibliometric analysis to examine assessment methods, technological innovations, and interdisciplinary approaches in STEM education. A total of 98 articles were retrieved from the Scopus database. Following the identification, screening, eligibility, and inclusion stages based on predefined inclusion and exclusion criteria, 36 articles published between 2019 and 2024 were selected for analysis. Bibliometric analysis examined keyword relationships and research networks, while VOS viewer visualized the bibliometric maps. The findings revealed that summative assessment was the dominant assessment method across educational levels, whereas formative assessment was also widely used to provide continuous feedback. Bibliometric analysis identified STEM education as the largest research cluster, closely related to technology integration and interdisciplinary learning. In addition, 12 of the 36 reviewed studies reported technology-enhanced and interdisciplinary assessment innovations, including eye-tracking, Bayesian inference, virtual robotics, game-based learning, and large language models. These findings highlight the potential of technology-enhanced and interdisciplinary assessment to improve STEM learning across educational levels.