Research

Psychometrics

Our lab is dedicated to advancing educational and psychological measurement by developing innovative quantitative methods to address emerging challenges in the field. We focus on pushing the boundaries of how learning and behavior are understood and measured.

Our research spans a range of areas, including change point analysis, categorical data analysis, item response theory, and cognitive diagnosis models. In addition, we are exploring the integration of machine learning and artificial intelligence to enhance these methods, enabling more accurate assessment, efficient data analysis, and the identification of key features for personalized learning and targeted interventions.

Through this work, we aim to contribute both to the theoretical foundations of psychometrics and to practical tools that improve educational and psychological research and practice.

Machine learning and AI technologies

Our lab is also deeply interested in exploring the applications of machine learning and artificial intelligence in education and psychology. We aim to leverage these advanced technologies to enhance research in areas such as cognitive diagnosis models, change-point detection, and the identification of key features for diagnosis and intervention.

By integrating AI and machine learning with traditional psychometric methods, we strive to improve the accuracy, efficiency, and interpretability of assessments, ultimately supporting more personalized learning and targeted psychological interventions.

Research Platform Design and Development

Another goal we have is to build an interdisciplinary research hub to support modern educational and psychological research, particularly to support the cyberinfrastructure design and development, which facilitates the research data collection and analysis.

Currently, the lab is working on the development of three research platforms:

Intelligent Diagnostic Assessment Platform (i-DAP) for High School Statistics Education.

i-DAP is an education platform funded by the NSF and IES, developed in collaboration with Prof. Alison Cheng’s research group. The platform aims to integrate state-of-the-art quantitative methods to address challenging assessment problems in both education and psychology. Over the years, it has been used by numerous students and has supported a wide range of research, resulting in many publications that address diverse research questions.

Automated LLM-assisted Item Generation Platform

The Automated LLM-assisted Item Generation Platform is a research platform designed to explore how modern large language models (LLMs) can be leveraged to facilitate the automated generation of assessment items. This platform aims to use AI to improve the efficiency, scalability, and quality of item development for educational and psychological assessments. The project is a collaborative effort with Adapta Education and is supported by the NSF’s STTR grant.

Computerized Adaptive Test for Suicide Risk Pathways (CAT-SRP)

CAT-SRP is an Ecological Momentary Assessment (EMA) platform integrated with Computer Adaptive Testing, designed to deliver timely and personalized mental health interventions. By combining real-time data collection with adaptive testing, the platform enables precise monitoring of psychological states and supports immediate, data-driven actions. This collaborative project, led by Prof. Brooke A. Ammerman and Prof. Ross Jacobucci, aims to advance research and practice in mental health assessment by providing more accurate and effective interventions.

Data analytics and modeling

Our lab also collaborates closely with other research entities on campus, including the Lucy Family Institute for Data and Society, the Institute for Educational Initiatives, and the Department of Computer Science, to support their data collection and analysis.