Description:

Yajuan Si, PhD

Research Assistant Professor
Survey Research Center
Institute for Social Research
University of Michigan, Ann Arbor

Presentation Title

Multilevel Regression and Poststratification: A Unified Framework for Survey Weighted Inference 

Objectives

Real-life sample data are often unrepresentative due to selection bias and nonresponse. Classical design-based corrections rely on weighting to match the sample to the target population. However, volatile weights can result in unstable estimators. We have developed a unified framework for survey weighting through novel modifications of the multilevel regression and poststratification (MRP). This talk will present a guide to the creation of survey weights using the Fragile Family and Child Wellbeing study as an example, and the implementation of MRP in electronic health records to estimate community-level SAR-COV-2 infection rates.