Appendix D
Biosketches of Contracted Project Staff
Dan Liao is a senior research statistician and program manager at RTI International who has focused on developing and implementing innovative techniques in statistics and data science. She has led on multiple statistical aspects for large national surveys in the U.S., including multi-phase and longitudinal sampling design, survey weighting, data editing and imputation, statistical disclosure control, and the analysis of complex survey data. Currently, she serves as the lead statistician in the National Longitudinal Study of Adolescent to Adult Health and the National Survey of Mental Health. Liao also pioneered in expanding official statistics by integrating survey science and statistical learning techniques with administrative data and data collected from other sources. Her statistical training is complemented with training on data-collection techniques, issues related to data-collection mode bias, nonresponse bias, experimental design, and questionnaire design. Liao has published research on regression diagnostics for complex survey data, calibration weighting for unit nonresponse, combining data from multiple sources, data quality improvement, Bayesian mixture modeling, and predictive modeling. She is also an active member and educator in the survey statistics community. She teaches a graduate-level course on applied regression at Columbia University’s Mailman School of Public Health. Her extensive professional service includes positions such as associate editor for Journal of Survey Statistics and Methodology, the methodology section chair for the Washington Statistical Society, and the publication officer for the Survey Research Methods Section of the American Statistical Association. Liao earned her Ph.D. in survey methodology from the University of Maryland.
Sahar Zangeneh is a research statistician at RTI International who specializes in various aspects of the statistical design and analysis of complex surveys, clinical trials, and observational studies. Zangeneh’s methodological interests include (1) statistical analysis with missing data, including non-ignorable missingness; (2) model-based analysis of complex survey data; and (3) integrating data from multiple sources. In her research, she uses classical and modern statistical methods, including frequentist and Bayesian methods, and machine-learning tools for high-dimensional data. She has received multiple awards for her methodological research from scientific communities such as the American Statistical Association, the International Society for Bayesian Statistics, and the Institute of Mathematical Statistics. Before joining RTI International, Zangeneh was a faculty member at Fred Hutchinson Cancer Research Center, where she performed methodological research; developed and implemented research protocols for diverse domestic and international health projects; and mentored graduate students, junior researchers, and statistical analysts. The depth of her statistical skills and the breadth of her collaborative experiences uniquely position her to conduct multidisciplinary research that answers a diverse array of scientific questions while preserving a high level of statistical rigor. Zangeneh holds faculty positions at University of Washington and Fred Hutchinson Cancer Research Center in Seattle. She is passionate about engaging in initiatives for fostering inclusion and diversity and promoting STEM fields in underrepresented and underserved students.
Jennifer J. Unangst is a research statistician at RTI International with more than eight years of experience performing and overseeing sample design, selection, and weighting for household surveys; statistical data processing and analysis; and programming for data editing and quality control in both SAS and SUDAAN. She has worked on more than 15 complex surveys and evaluation studies for a variety of clients such as the U.S. Department of Labor and the National Center for Health Statistics. Currently, she serves as the statistical task leader for O*NET, an establishment-based survey of employees in the U.S. designed to characterize the nation’s occupations. She has published on topics including gridded population-sampling methods, data-integration methods, and investigating error sources present in survey data. Prior experience investigating data quality includes a feasibility study for the U. S. assistant secretary for planning and evaluation; this study assessed some of the error sources present in disability, health, and caregiving data collected using internet panels compared to traditional surveys. Unangst earned a master’s degree in statistics from North Carolina State University and a B.A. in mathematics from West Chester University.
John David Bunker, Jr. is a statistician at RTI International with more than five years of professional experience in statistical modeling, imputation, sample weighting, and machine learning. His research interests include outlier detection for administrative data, unsupervised clustering, and variance and bias estimation for nonprobability samples. His most recent project experience includes the National Incident-Based Reporting System; Rigorous Evaluation of Pharmacy Strategies to Address Heart Disease and Stroke Prevention; Saving for Education, Entrepreneurship, and Downpayment for Oklahoma Kids; and various other projects where he provides extensive support in statistics and data science. Bunker earned an M.A. in statistics from the University of California, Berkeley, and a B.S. degree in mathematical decision sciences as well as a B.S. in psychology from the University of North Carolina at Chapel Hill.
Philip Lee is a research statistician at RTI International with eight years of experience working with complex data in both survey and administrative data contexts. He has experience in imputation and weighting of survey data, sampling, data analysis, and editing, and is proficient in SAS, R, and VBA programming. Currently, he is a key statistician for the NCS-X estimation project, in which he focuses on performing multilevel imputation and creating estimates at national and subnational levels based on data collected from the Federal Bureau of Investigation’s National Incident-Based Reporting System, which has become the U.S.’s primary crime reporting system since 2021. He also provides statistical and programming supports for large-scale national surveys, such as the National Study on Drug Use and Health, the National Surveys of Child and Adolescent Well-Being, and the National Crime Victimization Survey. Lee received a B.S. in mathematics and a B.A. in economics from the University of Maryland.
This page intentionally left blank.