Teaching regression calibration to correct for measurement error to develop statistical thinking
DOI:
https://doi.org/10.52041/iase25.144Abstract
Introductory statistics and biostatistics courses are essential for cultivating statistical thinking. This work explores the effectiveness of teaching measurement error (ME) methods in an introductory biostatistics course to teach three statistical concepts: bias, uncertainty, and decision-making. A tutorial was developed to illustrate the application of ME methods for a specific case study using a sample of U.S. adults (n=600) from the National Health and Nutrition Examination Survey (NHANES). Data were analyzed using logistic regression models to predict 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, simulated random noise was introduced based on reliability levels. Regression calibration (RC) was applied and improved the estimate of the noisy data. Students were taught that RC may reduce the effect of ME in statistical modeling to reduce bias, and to also improve uncertainty and decision-making. This work demonstrated the utility of an ME tutorial to develop understanding of statistical bias, uncertainty, and decision-making.References
Jaeger, B. C., Chen, L., Foti, K., Hardy, S. T., Bress, A. P., Kane, S. P., Huang, L., Herrick, J. S., Derington, C. G., Poudel, B., Christenson, A., Colantonio, L. D., & Muntner, P. (2023). Hypertension Statistics for US Adults: An Open-Source Web Application for Analysis and Visualization of National Health and Nutrition Examination Survey Data. Hypertension, 80(6), 1311–1320. https://doi.org/10.1161/hypertensionaha.123.20900
Murillo A. L. & Maytin, A. (2024). Tutorial of Measurement Error Methods for Evaluating
Blood Pressure using NHANES Data [R Software]. GitHub.
https://anarinalm.github.io/BP_MeasurementError/visualization.html Prentice R. L. (1996). Measurement error and results from analytic epidemiology: dietary fat and breast cancer. Journal of the National Cancer Institute, 88(23), 1738–1747. https://doi.org/10.1093/jnci/88.23.1738
Rumsey, D. J. (2002). Statistical literacy as a goal for introductory statistics courses. Journal of statistics education, 10(3). https://doi.org/10.1080/10691898.2002.11910678
Varlamis, I. (2025). Messy Data in Education: Enhancing Data Science Literacy Through Real- World Datasets in a Master’s Program. Education Sciences, 15(4), Article 500. https://doi.org/10.3390/educsci15040500
