Abstract
Much of the data in social and biomedical sciences features non-ignorable correlations that must be considered in analysis. This correlation arises when multiple observations are taken on the same subject or cluster of subjects (e.g., classroom, child development center), and/or near one another in space, time or jointly in space time. Depending on the modeling goals, the analyst must make several consequential statistical choices that are often under-discussed.
This presentation will cover several projects rooted in education and public health that feature correlated data, the methodological hurdles and triumphs therein. It will also discuss some opportunities to extend existing models, such as areal indices (e.g. ADI, COI) using spatial models; educational assessment data using Item Response Models; and displayed effect from automatic facial coding software.
Details
Date, Time & Location
April 24, 2026
10:00-11:30 A.M. CDT
Location TBD
This presentation is free and open to the public. It is made possible by support from the NU Nebraska Research Initiative.
Pavel Chernyavskiy, Ph.D.
Assistant Professor, Department of Health Sciences, University of Virginia
Pavel Chernyavskiy's methodological work is motivated by the analysis of correlated data, broadly construed. Correlated data arise from taking multiple observations on the same subject or cluster of subjects, taking observations near one another in space, in time, or jointly in space and time. His areas of collaborative research have spanned public health, education, psychology and ecology.
His primary expertise is in spatial statistics, hierarchical models, Bayesian statistics and statistics education. His research interests also focus on disparities, population health, non-stationary spatial correlation and intervention efficacy.
Prior to joining the University of Virginia in January 2021, Chernyavskiy served as an assistant professor of statistics at the University of Wyoming (2018-2020) and a postdoctoral fellow at the National Cancer Institute–Division of Cancer Epidemiology and Genetics (2015-2018). He earned his Ph.D. in statistics from the University of Nebraska-Lincoln in 2015.
Full Biography