Research Statement

My research centers on the development and application of advanced quantitative methodologies to address pressing questions in education and public health. I am particularly interested in how data science can reveal hidden patterns in complex systems, inform decision-making, and improve outcomes for underserved populations. My methodological expertise includes latent variable modeling, regression analysis, propensity score matching, multilevel modeling, and social network analysis, using tools such as R, Python, SPSS, and SQL.

Education Data Science

My work in education data science explores how student learning behaviors, instructional practices, and contextual factors interact to influence academic outcomes. I apply machine learning, educational data mining, and psychometric modeling to analyze large-scale student data and improve educational equity.

One line of research investigates the alignment between instructional content and student writing using topic modeling (e.g., LDA, BERTopic) and text classification techniques. These methods allow me to detect off-topic or low-effort responses and identify misalignment between student submissions and instructional objectives. In other work, I have used time-series analysis and clustering to track student engagement patterns on learning platforms, with the goal of predicting dropout risk and tailoring interventions in real-time.

I also explore measurement issues in education, particularly around measurement inequivalence in assessments. My dissertation examined how psychological instruments may function differently across subpopulations and applied methods to balance this inequivalence to improve validity and fairness. This work is grounded in a justice-oriented perspective that seeks to make invisible patterns—particularly those affecting marginalized or underrepresented learners—visible through rigorous analysis.

In collaborative projects, I have used structural equation modeling to study the relationship between teacher preparation, classroom design, and student outcomes. My role typically includes database management, instrument design, and multivariate modeling, with findings presented at national conferences such as AERA and APHA.

Public Health & Healthcare Analytics

My research in public health and healthcare analytics applies advanced statistical and machine learning techniques to real-world datasets focused on behavioral health, homelessness, and recovery. I have contributed to multiple federally funded projects, including those supported by SAMHSA, where I led data collection, predictive modeling, and evaluation design.

A key focus has been improving services for individuals experiencing homelessness and substance use disorders. I conducted hierarchical linear modeling and social network analysis to evaluate community-based interventions, often working closely with service providers to translate findings into actionable insights. I have also led predictive modeling efforts to identify recovery factors and risk indicators in large longitudinal datasets, helping inform the design of more responsive care systems.

My public health research is interdisciplinary in nature, often involving collaborations across social work, clinical research, and health policy. I prioritize transparency, reproducibility, and ethical data use—principles I emphasize in both my academic writing and when mentoring students.

Future Directions

In both education and public health, I plan to continue integrating computational approaches such as natural language processing and AI-assisted learning analytics. My long-term goal is to contribute to cross-sector efforts that leverage data to reduce disparities and inform evidence-based policy. I also aim to mentor students in applied research settings, empowering the next generation of data-informed scholars and practitioners.