The Survey Research Methods Section is sponsoring a continuing education course at the 2025 Joint Statistical Meetings titled “Statistical Tools for Analysis of Nonprobability Samples.” It will be taught by Jae-Kwang Kim from Iowa State University and Yonghyun Kwon from the Korea Military Academy.
The course will cover state-of-the-art tools for analyzing nonprobability samples, focusing on the statistical theory and methods for propensity score weighting. Included will be a demonstration using real-life data examples and relevant R packages.
Participants should have basic knowledge of mathematical statistics and survey sampling (undergraduate level) and some experience with R programming.
The SRMS webinar series aims to provide a platform for survey researchers to share their knowledge on survey research methods and for participants to learn from the speakers and exchange ideas. Last year, Yajuan Si presented “A Comparative Review of Data Integration Methods.” The section also sponsored a career advice webinar on opportunities in industry.
Applications to deliver webinars on cutting-edge topics such as nonprobability samples, machine learning, and AI techniques are wanted. Contact Monika Hu if you are interested in presenting in the webinar series. Both members and nonmembers are welcome to submit ideas.
Name: Jim Knaub
Mail: jamesRknaub@gmail.com
Website: https://www.researchgate.net/profile/James-Knaub
“R packages” were noted and I will add that one to consider for nonprobability sampling is found at https://github.com/ncn-foreigners/nonprobsvy. There it says:”nonprobsvy: an R package for modern statistical inference methods based on non-probability samples.” That package is under construction. You may install a recent version or a development version. It deals with “inverse probability weighting estimators,” “mass imputation estimators,” and “double robust estimators.” Further, a possible future ‘enhancement,’ found under “issues,” #78, would cover “Cutoff or Near-Cutoff Sampling with Prediction,” methodology which has been highly productive at the US Energy Information Administration.