My research addresses fundamental questions about educational measurement and assessment in small-sample, many-group contexts:
How can we design assessments and models that reliably measure student outcomes and support equitable decision-making when data are limited and highly segmented?
This question sits at the heart of my work and connects issues of psychometric modeling, predictive analytics, and educational policy.Research Areas
Psychometric Theory and Methods
My methodological research focuses on statistical models tailored to small-sample, multi-group data.
Item Response Theory (IRT)
Models for understanding item characteristics and student ability across many groups.
Models for understanding item characteristics and student ability across many groups.
Bayesian Methods
Using priors and hierarchical models to improve estimation with limited data.
Using priors and hierarchical models to improve estimation with limited data.
Neural Networks for DIF
Multilabel classification approaches to detect differential item functioning when sample sizes are small.
Multilabel classification approaches to detect differential item functioning when sample sizes are small.
Educational Data Science
I study how predictive modeling and computational methods can improve educational measurement.
Machine learning for identifying at-risk students in highly segmented populations.
Combining psychometric and computational approaches to support equity and generalizability.
Evaluating the trade-offs between frequentist and Bayesian approaches for sparse response data.
Applied Impact
My applied work integrates research with collaboration to influence educational practice and policy.
Designing dashboards and visualizations to guide equitable resource allocation.
Supporting educators and policymakers in using assessment data responsibly.
Consulting on research design, statistical analysis, and evidence-based decision-making.
Current Work:
I am currently exploring methods to improve measurement and modeling in highly segmented or sparse data:- Developing generalizable methods detect small sample differential item functioning across many groups.
- Using machine learning and Bayesian methods for Q-Matrix estimation and refinement, to enhance measurement precision.
- Comparing frequentist and Bayesian approaches for modeling sparse responses, to balance data simplification with principled prior modeling.