Colloquium: Teaching Measurement Error Models in Introductory Biostatistics Courses to Develop Statistical Thinking

Teaching Measurement Error Models in Introductory Biostatistics Courses to Develop Statistical Thinking
Anarina Murillo, PhD, Assistant Professor of Biostatistics, Brown University Tuesday, March 4, 2025, at 11 AM in the ZSR Auditorium

Abstract: Introductory biostatistics courses are essential for teaching statistical thinking and data literacy. A problem-first approach allows students to draw connections in statistics and data science to their respective fields in medicine or public health. Novel approaches are needed to introduce advanced statistics concepts in these courses to foster critical thinking using real-world applications. This work explores the effectiveness of teaching measurement error (ME) methods in an introductory biostatistics course to develop understanding of bias, uncertainty, and decision-making. A ME model tutorial was developed using the National Health and Nutrition Examination Survey (NHANES) data. We proposed a logistic regression model to predict adult diabetes status based on blood pressure levels adjusted for age, race, calories, alcohol intake, and body mass index (BMI). To imitate real world applications of ME correction, we analyzed a subset of the data (n=600) and simulated random noise to imitate random error in our dataset. Regression calibration (RC) was applied to correct for the simulated error in the model. Students were taught that RC may reduce the effect of ME in statistical models. Students reported improved understanding of statistical inference. This work demonstrated the utility of a web-based tutorial on RC to introduce advanced statistical topics in an introductory course to deepen understanding of bias, uncertainty, and decision-making.

Anarina Murillo, PhD, is an Assistant Professor of Biostatistics on the Teaching Scholar Track at Brown University. Professor Murillo received her PhD in Applied Mathematics and a Certificate in Statistics from Arizona State University. She completed an NIH T32 postdoctoral fellowship in the statistical genetics and obesity training programs in the Department of Biostatistics and NIH-funded Nutrition Obesity Research Center (NORC) at the University of Alabama at Birmingham. She served as Visiting Faculty in the Department of Biostatistics at the NYU School of Global Public Health. Prior to that she was an Assistant Professor (Research) with the Department of Pediatrics and a Senior Biostatistician with the Center for Statistical Sciences at Brown. She was involved with leading the design and analysis of clinical and translational research in collaboration with investigators affiliated with the Advance-CTR, Hasbro Children’s Hospital, and Women & Infants Hospital.In 2020, she was selected for the NIH NHLBI PRIDE program as a Research in Implementation Science for Equity (RISE) fellow at the University of California, San Francisco. In 2022, she was selected for the University of Pennsylvania’s Center for Improving Care Delivery for the Aging (CICADA) Scholars program.