Dr. Ummugul Bezirhan
Assistant Research Director for Advanced Psychometrics and Machine Learning Applications

Dr. Ummugul Bezirhan serves as Assistant Research Director for Advanced Psychometrics and Machine Learning Applications. In this role, Dr. Bezirhan leads research and development initiatives at the intersection of psychometrics, artificial intelligence (AI), and large-scale assessment. Her work spans multiple stages of the assessment cycle, including automated scoring, item generation, item classification, and test assembly. In particular, Dr. Bezirhan is advancing the field through pioneering work on LLM-based automated scoring and embedding-based approaches to reliability auditing, leveraging natural language processing to enhance the precision and robustness of score interpretations in large-scale assessments.

Dr. Bezirhan plays a central role in integrating AI-driven approaches with established psychometric methods. She also contributes to international reporting, secondary data analysis, and capacity building around the ethical and effective use of AI in educational measurement. In addition to her research contributions, Dr. Bezirhan is responsible for operational psychometric scaling for TIMSS, PIRLS, and other special studies. She has played a key role in the Center’s efforts to streamline and modernize scaling workflows and in developing and implementing updated psychometric scaling procedures.

Dr. Bezirhan holds a PhD in Measurement, Evaluation, and Applied Statistics from Columbia University. Her dissertation received the Dissertation Award from the Psychometric Society. Prior to joining the TIMSS & PIRLS International Study Center, she held research and data science positions at the National Board of Medical Examiners (NBME), the American Institutes for Research (AIR), and NBCUniversal.