My research is at the intersection of AI, education, psychology, and technology. Specifically, my research is centered around:

  • Developing and applying educational data mining and learning analytics methods on process data to explore how learners interact with technology-based learning and assessment systems, and how these interactions relate to cognition and learning,
  • Ensuring fairness in AI and data science methods used in education, and
  • Assessing and supporting complex 21st-century skills (e.g., collaborative problem solving, self-regulated learning).

Educational Data Mining and Learning Analytics on Process Data

I am passionate about developing and applying educational data mining and learning analytics methods on process data to explore how learners interact with technology-based assessments and learning systems, and how the interactions are related to cognition and learning. My work often involves analyzing the wealth of process data (e.g., log files, text chats, audio, video) obtained from various types of systems, including large-scale educational assessments, intelligent tutoring systems, simulations, educational games, and virtual environments. For example, in a series of studies using data from NAEP’s digitally based assessments, we investigate test-takers’ cognitive and metacognitive processes and problem-solving strategies by integrating process data, response data, and survey questionnaire data. To uncover patterns in learner behavior and cognition, I apply advanced methods including sequence mining, feature engineering, cluster analysis, social network analysis, hidden Markov models, and visualization techniques like Sankey diagrams to make visible what is often hidden in traditional assessments. Below are select projects related to this line of work.

Automated Writing Feedback JCAL
Unveiling Patterns of Interaction with Automated Feedback in Writing Mentor and Their Relationships with Use Goals and Writing Outcomes
Yang Jiang, Beata Beigman Klebanov, Jiangang Hao, Paul Deane, Oren E. Livne
Journal of Computer Assisted Learning, 2025.
NAEP Calculator Use COMP EDU
Using Sequence Mining to Study Students’ Calculator Use, Problem Solving, and Mathematics Achievement in the National Assessment of Educational Progress (NAEP)
Yang Jiang, Gabrielle A. Cayton-Hodges, Leslie Nabors Oláh, Ilona Minchuk
Computers and Education, 2023.
Data Mining and Problem Solving JRME
Investigating Problem Solving on Calculator Items in a Large-Scale Digitally-Based Assessment: A Data Mining Approach
Yang Jiang, Gabrielle A. Cayton-Hodges
Journal for Research in Mathematics Education, 2023.
Process Data LSAE
Using Process Data to Understand Problem-Solving Strategies and Processes for Drag-And-Drop Items in a Large-Scale Mathematics Assessment
Yang Jiang, Tao Gong, Luis E. Saldivia, Gabrielle Cayton-Hodges, Chris Agard
Large-Scale Assessments in Education, 2021.
Deep Learning AIED
Expert Feature-Engineering vs. Deep Neural Networks: Which is Better for Sensor-Free Affect Detection?
Yang Jiang, Nigel Bosch, Ryan S. Baker, Luc Paquette, Jaclyn Ocumpaugh, Juliana Ma Alexandra L. Andres, Allison L. Moore, Gautam Biswas
Proceedings of the 19th International Conference on Artificial Intelligence in Education (AIED 2018)
🏆 [Best Student Paper Award] [Nominated for Best Paper Award]

Responsible and Ethical Use of AI in Education

The rapid rise of generative AI, such as ChatGPT, presents both tremendous opportunities and complex challenges. One significant challenge for educators is determining whether students’ submissions are their own work or AI-generated—a task crucial for ensuring academic integrity. To address this, we developed and evaluated various methods for detecting AI-generated essays by leveraging extensive data from large-scale assessments. We systematically investigated detectors’ performance across demographic groups, ensuring the detectors are fair with no systemic disadvantages for marginalized students. Our detectors achieved near-perfect accuracy and minimal bias against non-native English speakers. Expanding this work, we incorporated process data such as keystroke dynamics and writing behaviors to identify nonauthentic writing behaviors, opening new avenues for enhancing detection methods while prioritizing fairness. Our findings offer empirical evidence to inform educational policies, support the responsible use of AI tools, and promote fairness in academic environments. This work was recognized with the Harvard Graduate School of Education Alumni Council Award for Impact in Education.

AI Detection COMP EDU
Detecting ChatGPT-Generated Essays in a Large-Scale Writing Assessment: Is There a Bias Against Non-Native English Speakers?
Yang Jiang, Jiangang Hao, Michael Fauss, Chen Li
Computers and Education, 2024.
Bias Study AIED
Towards Fair Detection of AI-Generated Essays in Large-Scale Writing Assessments
Yang Jiang, Jiangang Hao, Michael Fauss, Chen Li
Proceedings of the 25th International Conference on Artificial Intelligence in Education (AIED 2024)
Keystroke Patterns JEM
Using Keystroke Behavior Patterns to Detect Nonauthentic Texts in Writing Assessments: Evaluating the Fairness of Predictive Models
Yang Jiang, Mo Zhang, Jiangang Hao, Paul Deane, Chen Li
Journal of Educational Measurement, 2024.

Assessment of Complex 21st Century Skills in the Age of AI

I have led and contributed to numerous large-scale, multi-year research projects focused on assessing and supporting complex 21st century skills, including collaborative problem solving (CPS) and self-regulated learning (SRL). These projects span a wide range of domains, task types, participant populations, and experimental designs, enabling us to examine the generalizability of our methodologies and findings across diverse contexts. We apply theoretically grounded approaches that integrate educational data mining, psychometric modeling, and data analytics on multimodal data to better measure these complex constructs and understand their relationship to learning outcomes. This research provides deeper insights into the dynamics of human-human and human-AI interactions, supports more valid assessments of complex skills, and informs the design of interventions or scaffolding to support the development and transfer of these critical skills.

LLM Coding AIED
Uncovering Transferable Collaboration Patterns Across Tasks Using Large Language Models
Yang Jiang, Jiangang Hao, Wenju Cui, Emily Kerzabi, Patrick Kyllonen
Proceedings of the 26th International Conference on Artificial Intelligence in Education (AIED 2025)
Epistemic Network Analysis ISLS
Using Epistemic Network Analysis and Sequential Pattern Mining to Explore the Impacts of Human Facilitation on Collaborative Mathematical Problem Solving
Yang Jiang, Edith Aurora Graf, Jessica Andrews-Todd
Proceedings of the Annual Meeting of the International Society of the Learning Sciences (ISLS 2025)
Personality CHB
Do You Know Your Partner's Personality Through Virtual Collaboration or Negotiation? Investigating perceptions of personality and their impacts on performance
Yang Jiang, Michelle Martín-Raugh, Zhitong Yang, Jiangang Hao, Lei Liu, Patrick C. Kyllonen
Computers in Human Behavior, 2023.
CPS Across Tasks COMP EDU
Investigating Collaborative Problem Solving Skills and Outcomes Across Computer-Based Tasks
Jessica Andrews-Todd, Yang Jiang, Jonathan Steinberg, Samuel L. Pugh, Sidney K. D’Mello
Computers and Education, 2023.
SRL in VPA Book Chapter
How Immersive Virtual Environments Foster Self-Regulated Learning
Yang Jiang, Jody Clarke-Midura, Ryan S. Baker, Luc Paquette, Bryan Keller
Digital Technologies and Instructional Design for Personalized Learning, 2018
Note-Taking in OELE CEP
Note-Taking and Science Inquiry in an Open-Ended Learning Environment
Yang Jiang, Jody Clarke-Midura, Bryan Keller, Ryan S. Baker, Luc Paquette, Jaclyn Ocumpaugh
Contemporary Educational Psychology, 2018