It is with great excitement that we introduce a new volume, Research for Practical Issues and Solutions in Computerized Multistage Testing, edited by Duanli Yan, David Weiss, and Alina von Davier.
When the first volume of this work, Multistage Adaptive Testing: Theory and Applications, was published 10 years ago, together with Charlie Lewis, adaptive multistage testing was emerging as a promising alternative to traditional assessment methods. Today, it has solidified its place as a robust, scalable approach used across diverse fields, including education, psychological testing, professional certification, and corporate training.
The rise of adaptive testing
One of the most significant changes in adaptive testing has been the advancement of technology. Computational psychometrics, machine learning and AI-driven algorithms have revolutionized item development, item and module selection, scoring, test security, and real-time data analysis, leading to more precise, secure, and efficient assessments.
Additionally, collecting new types of data—such as response times and process data—at scale has enhanced our understanding of test-takers’ abilities and learning processes, making adaptive testing an even more powerful tool for educational and psychological measurement.
Beyond technological improvements, adaptive testing has also gained wider adoption globally. Its ability to balance flexibility and fairness in large-scale assessments has made it an attractive option for organizations seeking reliable, adaptive testing solutions.
MST in action: The Duolingo English Test
One of the most notable real-world examples of how adaptive testing methodologies have evolved is the Duolingo English Test (DET), which exemplifies the advances made in adaptive tests and AI-driven assessments over the past decade.
Unlike traditional language proficiency exams, the DET leverages AI and adaptive testing to dynamically adjust the difficulty of questions based on a test-taker’s responses in real-time. This approach ensures a highly efficient and personalized assessment experience, reducing test length while maintaining measurement accuracy.
Additionally, the DET incorporates multimodal data, such as speech analysis and behavioral metrics, to enhance scoring and security. By making high-quality language assessment accessible, scalable, and equitable, the DET reflects the transformative power of technology in modern testing—a vision that aligns with the ongoing advancements explored in this new volume, Research for Practical Issues and Solutions in Computerized Multistage Testing.
Revisiting foundational theories while looking ahead
This new volume revisits foundational theories and applications of adaptive multistage testing. It also addresses recent innovations and challenges that have come to light as the field has evolved over the past decade.
Contributions from leading researchers provide insights into how MST continues to evolve in response to technological advancements and changing assessment needs. New types of assessment designs, such as the personalized ensemble test design, hybrid designs, or entangled test designs are now possible—at scale, for high-stakes tests.
Importantly, this edition also tackles the ethical considerations associated with AI-driven testing. As adaptive testing systems become more sophisticated, concerns around transparency, fairness, and bias have grown. Ensuring that adaptive testing remains equitable for all test-takers requires careful design, ongoing research, and collaboration among experts in assessment, AI engineering, and education.
The power of collaboration in scientific progress
Scientific progress thrives on collaboration, and the development of MST is no exception. The editors of both volumes—spanning decades of friendship and professional camaraderie—exemplify how collaboration fuels innovation. This volume is a testament to the collective efforts of researchers who have contributed their expertise to advancing the science of adaptive testing.
Professional organizations like the International Association of Computerized Adaptive Testing (IACAT) have played a vital role in fostering this collaborative spirit. Over the past ten years, IACAT and similar organizations have provided platforms for knowledge-sharing through conferences, workshops, and publications. These gatherings are more than just academic events; they serve as incubators for new ideas, partnerships, and research that continue to push the field forward.
A tribute to the past, a guide for the future
As we celebrate the ten-year milestone of Multistage Adaptive Testing: Theory and Applications, we recognize the contributions of the original editors and contributors who laid the groundwork for this evolving field. This new volume serves as both a tribute to their work and a forward-looking guide for researchers, practitioners, and policymakers dedicated to advancing MST.
The journey of MST is far from over. With continuous advancements in AI, data science, and assessment design, the next decade promises even greater innovations. As we reflect on the progress made, we look forward to how adaptive testing will continue shaping the future of assessment, ensuring more effective, fair, and adaptive testing for learners and professionals worldwide.
📄Check out more of Alina’s research contributions here!