Research Spotlight: Yiqin Pan
Q & A with Yiqin Pan, Ph. D., Assistant Professor 
in the School of Human Development and Organizational Studies in Education
What research are you currently working on?
I leverage quantitative methodologies, including artificial intelligence, statistical modeling, and psychometrics, to address applied issues in educational measurement and to optimize the learning process. Most of my recent research has centered on aspects of test security and personalized learning. My current projects include (i) developing anomaly detection algorithms for identifying potential fraud in tests, (ii) implementing item selection designs by recommendation systems for preventing potential fraud in adaptive testing, (iii) using anomaly detection methods to identify disengagement in learning, and (iv) building recommendation systems to select appropriate learning materials for students. My research has been supported by the Educational Testing Service (ETS) and the Graduate Management Admission Council (GMAC).
What is the broader impact of your research?
My research in test security aims to protect fairness in educational testing. Testing programs should give test takers a fair chance to demonstrate their abilities. However, cheating will provide an unfair advantage to cheaters, and has become one of the most common and significant concerns within the testing industry. To maintain the integrity of testing, my research has developed several strategies to mitigate the potential impact of cheating and several algorithms to detect cheaters after the fact. These methods can assist in protecting the fairness of the tests.
My research in personalized learning is intended to promote equity in education. Minority groups, low-income people, and women have long been underrepresented in STEM education. Such underrepresentation occurs because underserved students experience higher failure rates than other students. Personalized learning is an educational approach that empowers students’ voices and choices. The responses from students advance our understanding of the interests and needs of each student. The guidance provided based on personalized understanding could enhance students’ engagement and efficiency in learning. Higher engagement and increased productivity can prevent underrepresented populations from underperforming in learning and, thus, promote equity in education.
What other research topics are you interested in?
I am interested in applying machine learning to analyze process data and am also curious about using big data techniques to develop scalable algorithms for the analysis of large-scale assessment data.
