Emerging Trends in Computer-Based Testing: Insights from OASYS and the Impact of Generative Item Models

This chapter provides a comprehensive exploration of e-assessment, delving into its multifaceted advantages and challenges. It begins by tracing the developmental path and critical insights gathered from the utilization of the OASYS (Online-Assessment SYStem) assessment platform, a cornerstone of educational practices in Luxembourg. Subsequently, the focus shifts towards a critical aspect prevalent in all e-assessments: the creation of test content. Regardless of the intended application, whether it is used for adaptive testing, formative assessments, or summative evaluations, the necessity of robust and psychometrically sound test content remains paramount. Within this context, the chapter illustrates the innovative approaches adopted by the Luxembourg Centre for Educational Testing (LUCET) in addressing this challenge. Specifically, it highlights the implementation of template-based, generative item models in a large-scale mathematics assessment conducted nationwide. Furthermore, the chapter explores the growing interest in generative artificial intelligence (AI) and its potential implications in this context. Through a nuanced examination of these themes, this chapter offers valuable insights into the current trends and future directions of e-assessment.