About Me

I am an assistant professor of research methods, measurement, and evaluation in the Neag School of Education at the University of Connecticut. As an educational methodologist, I am committed to identifying effective educational programs and policies. I specialize in leveraging AI to extract insights from complex or hard-to-access data, including unstructured data on program implementation (like transcripts and participant writing). In doing so, I address critical resource constraints in education research while simultaneously addressing risks in the use and interpretation of AI output.

In recent research, I have:

• Developed a research protocol for increasing the validity of machine learning applications in education research

• Developed scalable methods for collecting local policy data from school district websites and used that data to estimate the impact of wide-scale deregulation

• Used large language models and natural language processing to analyze over 40,000 school board meeting minutes to assess the frequency of race-related discussions

• Demonstrated an effective approach to maximizing performance when using large language models to identify complex constructs in psychology

• Considered key validity concerns in the application of large language models to the Narrative Policy Framework

• Provided a tutorial on zero and few-shot classification, with a focus on validation

• Developed methods for measuring fidelity in standardized educational interventions

• Evaluated the causal validity of difference-in-differences and comparative interrupted time series designs, with an emphasis on the usefulness of common validity checks

You can see a complete list of my publications here.