Online Interdisciplinary Seminars on SM-SBR

Online Interdisciplinary Seminars on Statistical Methodology for Social and Behavioral Research

The online interdisciplinary seminars on statistical methodology for social and behavioral research is supported by the department of statistics and the department of education psychology in the University of Connecticut (UCONN), the Statistical and Applied Mathematical Sciences Institute (SAMSI) and the New England Statistical Society (NESS). The seminar is held online via WebEx and anyone in the world can join and it is scheduled monthly on Friday noon. The aims of the seminar are to promote the connection between the statistics and social/behavioral science communities and to encourage more graduate students to participate in the interdisciplinary research.

The past and current upcoming seminars are

11/20/2020 11:30-1:00pm  EST Bengt Muthen University of California Recent Advances in Latent Variable Modeling
12/18/2020 12:00-1:15pm  EST Paul De Boeck The Ohio State University Response Accuracy and Response Time in Cognitive Tests
1/29/2021 12:00-1:15pm  EST P. Richard Hahn Arizona State University The Bayesian causal forest model: regularization, confounding, and heterogeneous effects
2/26/2021 12:00-1:15pm  EST Edward Ip Wake Forest University Partially Ordered Responses and Applications
3/26/2021 12:00-1:15pm  EDT David Dunson Duke University Bayesian Pyramids: Identifying Interpretable Deep Structure Underlying High-dimensional Data
4/16/2021 12:00-1:15pm  EDT Susan Paddock NORC at the University of Chicago Causal Inference Under Interference in Dynamic Therapy Group Studies
4/23/2021 2:00-3:15pm  EDT Jean-Paul Fox University of Twente Bayesian Covariance Structure Modeling: An Overview and New Developments
4/30/2021 12:00-1:15pm  EDT Jennifer Hill  Columbia University  thinkCausal: One Stop Shopping for Answering your Causal Inference Questions
5/21/2021 12:00-1:15pm  EDT David Kaplan University of Wisconsin – Madison Developments and Extensions in the Quantification of Model Uncertainty: A Bayesian Perspective
6/18/2021 12:00-1:15pm  EDT Jon Krosnick Stanford University
9/10/2021 12:00-1:00pm  EDT Susan Murphy Harvard University
10/1/2021 12:00-1:00pm  EDT Fan Li  Duke University 
11/5/2021 12:00-1:00pm  EDT Jerry Reiter  Duke University

For announcements and WebEx live streaming links, please contact Tracy Burke (

For questions related to the seminars, please feel free to contact organizers
(Prof. Xiaojing Wang ( and Prof. Betsy McCoach ( ).

Bengt Muthen, Professor Emeritus, University of California, Los Angeles

Friday, November 20, 2020 11:30am -1:00pm

Recent Advances in Latent Variable Modeling

This talk gives an overview of some recent and ongoing latent variable research. Borrowing ideas from multilevel factor analysis, longitudinal SEM in a single-level, wide format is formulated in a new way that finds a well-fitting model 45 years after the writing of the classic Wheaton, Muthen, Alwin, and Summers article. This segues into a generalization of latent transition analysis using the multilevel notion of a random intercept while staying in a single-level, wide format. Turning back to multilevel modeling, the talk considers time series analysis of intensive longitudinal data. This is illustrated by intervention data on electricity consumption and a randomized intervention related to positive and negative affect where cycles play a major role. Finally, the new feature in Mplus Version 8.5 of Bayesian analysis of count, nominal, and binary logit models is presented.

This session is jointly sponsored by the Statistics department and the Research Methods, Measurement, and Evaluation program as part of the Statistical Applications and Quantitative Research Methods colloquium series.

Paul De Boeck , The Ohio State University

Friday, December 18, 2020 12:00 EST

Response Accuracy and Response Time in Cognitive Tests

It is an old and still unresolved issue how much a cognitive test score reflects ability and how much it reflects speed. The well-known speed-accuracy tradeoff does not make an answer to the question easier. In the presentation I will report the results of my research steps to investigate the problem. Briefly summarized, the findings are as follows. First, the correlation of ability and speed across persons depends on the test. Second, based on different kinds of modeling and different kinds of data, there seem to be remaining item-wise dependencies (i.e., conditional dependencies) between response accuracy and response time after controlling for the underlying latent variables. Third, the remaining dependencies depend on the difficulties of the test items and the dependencies also are curvilinear. I will present an explanation for the findings, and a tentative, complex answer to the old question what is being measured in a cognitive test.
This session is jointly sponsored by the Statistics department and the Research Methods, Measurement, and Evaluation program, University of Connecticut (UCONN), New England Statistical Society (NESS) and Statistical and Applied Mathematical Institute (SAMSI) as part of online interdisciplinary seminar series on statistical methodology for social and behavioral research.

P. Richard Hahn, Arizona State University

January 29, 2021- 12:00-1:15pm

The Bayesian causal forest model: regularization, confounding, and heterogeneous effects

In this talk, I will describe recent work on Bayesian supervised learning for conditional average treatment effects. I will motivate the proposed Bayesian causal forest model in terms of fixing two specific flaws with previous approaches. One, our model allows for direct regularization of the treatment effect function, providing lower variance estimates of heterogeneous treatment effects. Two, by including an estimate of the propensity score as a control variable in our model we mitigate a phenomenon called “regularization induced confounding” that leads to substantial bias in previous approaches. I will conclude with a detailed discussion of designing simulation studies to systematically investigate and validate machine learning models for causal inference.


Note: Dr. Hahn may also talk about this tutorial a bit:


Bio: Professor P. Richard Hahn has a B.A. in Philosophy of Science from Columbia University and earned his PhD in Statistics from Duke University in 2011. He taught at University of Chicago Booth School of Business for seven years before joining the School of Mathematical and Statistical Sciences at Arizona State University in 2017. His research lies at the intersection of machine learning and causal inference, specifically tree based regression methods for estimating heterogeneous treatment effects. Other research interests include latent variable models and statistical decision theory. He enjoys road trips in the mountain southwest with his family and riding and working on bicycles.


Dr. Edward Ip, Wake Forest School of Medicine

Friday, 2/26/2021, 12pm

Partially Ordered Responses and Applications

Partially ordered set (poset) responses are prevalent in fields such as psychology, education, and health. For example, the psychopathologic classification of no anxiety (NA), mild anxiety (MA), anxiety with depression (AwD), and severe anxiety (SA) form a poset. Due in part to the lack of analytic tools, poset responses are often collapsed into other data forms such as ordinal data. During such a process, subtle information within a poset is inevitably lost. In this presentation, a longitudinal latent-variable model for poset responses and its application to health data will be described. It is argued that latent variable modeling enables the integration of information from both ordinal and nominal components in a poset. Using the abovementioned example, NA>{MA,AwD}>SA form the ordinal component, and MA and AwD form the nominal component. Specifically, it will be demonstrated that the latent variable model “discovers” implicit ordering within the nominal categories. This is possible because both intra-person and inter-person information are borrowed to reinforce inference. Some potential applications of the poset model will also be highlighted.


Bio: Dr. Edward Ip is a Professor in the Department of Biostatistics and Data Science, in the Wake Forest School of Medicine. He received his master’s in education and PhD in statistics, both from Stanford. His research interests include latent variable modeling and longitudinal data analysis. He is currently Editor of the journal, Psychometrika, Application Reviews and Case Studies (ARCS) section.


Full WEBEX Info for this talk: (Friday, Feb 26, 2021 12:00 pm)

Meeting number: 120 286 8603

Password: RMMESTAT

Join by video system


You can also dial and enter your meeting number.
Join by phone

Join by phone

+1-415-655-0002 US Toll

Access code: 120 286 8603

Dr. David Dunson, Duke University

Friday, 3/26/2021, 12pm

Bayesian Pyramids: Identifying Interpretable Deep Structure Underlying High-dimensional Data

High-dimensional categorical data are routinely collected in biomedical and social sciences. It is of great importance to build interpretable models that perform dimension reduction and uncover meaningful latent structures from such discrete data. Identifiability is a fundamental requirement for valid modeling and inference in such scenarios yet is challenging to address when there are complex latent structures. We propose a class of interpretable discrete latent structure models for discrete data and develop a general identifiability theory. Our theory is applicable to various types of latent structures, ranging from a single latent variable to deep layers of latent variables organized in a sparse graph (termed a Bayesian pyramid). The proposed identifiability conditions can ensure Bayesian posterior consistency under suitable priors. As an illustration, we consider the two-latent layer model and propose a Bayesian shrinkage estimation approach. Simulation results for this model corroborate identifiability and estimability of the model parameters. Applications of the methodology to DNA nucleotide sequence data uncover discrete latent features that are both interpretable and highly predictive of sequence types. The proposed framework provides a recipe for interpretable unsupervised learning of discrete data and can be a useful alternative to popular machine learning methods.


Bio: Dr. David Dunson is Arts & Sciences Distinguished Professor of Statistical Science and Mathematics at Duke University. His research focuses on developing methodology for analysis and interpretation of complex and high-dimensional data, with a particular emphasis on biomedical applications, Bayesian statistics, and probability modeling approaches. Methods development and theory is directly motivated by challenging applications in neuroscience, genomics, environmental health, and ecology, among others. Dr. Dunson received his BS in Mathematics from the Pennsylvania State University in 1994, and his PhD in Biostatistics from Emory University in 1997. He then spent a decade at the National Institute of Environmental Health Sciences before moving to Duke. His work has had substantial impact, with ~55,000 citations on Google Scholar and an H-index of 80.


Full WEBEX Info for this talk: (Friday, March 26, 2021 12:00 pm)

Meeting number: 120 821 2088

Password: RMMESTAT

Join by video system

You can also dial and enter your meeting number.
Join by phone

Join by phone

+1-415-655-0002 US Toll

Access code: 120 821 2088

Dr. Susan Paddock, NORC University of Chicago

Friday, 4/16/2021, 12pm

Causal Inference Under Interference in Dynamic Therapy Group Studies

Group therapy is a common treatment modality for behavioral health conditions. Patients often enter and exit groups on an ongoing basis, leading to dynamic therapy groups. Examining the effect of high versus low session attendance on patient outcomes is of interest. However, there are several challenges to identifying causal effects in this setting, including the lack of randomization, interference among patients, and the interrelatedness of patient participation. Dynamic therapy groups motivate a unique causal inference scenario, as the treatment statuses are completely defined by the patient attendance record for the therapy session, which is also the structure inducing interference. We adopt the Rubin Causal Model framework to define the causal effect of high versus low session attendance of group therapy at both the individual patient and peer levels. We propose a strategy to identify individual, peer, and total effects of high attendance versus low attendance on patient outcomes by the prognostic score stratification. We examine performance of our approach via simulation, apply it to data from a group cognitive behavioral therapy trial for reducing depressive symptoms among patients in a substance use disorders treatment setting, and discuss the strengths and limitations of this approach.


Bio: Dr. Susan Paddock is the chief statistician and executive vice president at NORC at the University of Chicago. She is responsible for the methods of design and analysis used in NORC proposals and projects and for the NORC corporate research and development enterprise. Her research includes developing innovative statistical methods, with a focus on Bayesian methods, multilevel modeling, nonparametric Bayes, longitudinal data analysis, and missing data techniques. Dr. Paddock is the principal investigator of a project sponsored by the National Institute on Alcohol Abuse and Alcoholism to develop methods for analyzing data arising from studies of group therapy-based interventions. She was the principal investigator of a project sponsored by the Agency for Healthcare Research and Quality to improve the science of public reporting of health care provider performance. She co-led a project to conduct analyses related to the Medicare Advantage Plan Ratings for Quality Bonus Payments. Her other substantive research interests include health services research, substance abuse treatment, quality of health care, and veterans’ health care. Prior to joining NORC, she spent 20 years as a senior statistician with RAND Corporation. From 2008 to 2013, she led the RAND Statistics Group. She has served on editorial boards for the Annals of Applied Statistics, Journal of the American Statistical Association, and Medical Care, and has served on committees for the American Statistical Association (ASA) and the National Academies of Sciences, Engineering, and Medicine. She is a fellow of the ASA and was the recipient of the 2013 Mid-career Award of the Health Policy Statistics Section of the ASA. She received her PhD in statistics from Duke University and her BA in mathematics and biostatistics from the University of Minnesota.


Full WEBEX Info for this talk: (Friday, April 16, 2021 12:00 pm)

Meeting number: 120 026 0002

Password: RMMESTAT

Join by video system

You can also dial and enter your meeting number.

Join by phone

+1-415-655-0002 US Toll

Access code: 120 026 0002

Dr. Jean-Paul Fox, University of Twente

Friday, 4/23/2021, 2pm

Bayesian Covariance Structure Modeling: An Overview and New Developments

There is large family of statistical models to understand clustered or hierarchical structures in the data (e.g., multilevel models, mixed effect models, random effect models). The general modeling technique is to use a latent variable (i.e., random effect, frailty parameter) to describe the covariance among clustered observations, where the strength of the covariance is represented by the latent variable variance. This approach has several disadvantages. It is only possible to describe positive within-cluster correlation (similarity), and not dissimilarity (Nielsen et al., 2021). Sample size restriction and model complexity are often implied by the number and type of latent variables. Furthermore, the latent variable variance is restricted to be positive, which leads to boundary issues at/around zero and statistical issues in evaluating data in support of a latent variable. A new approach for modeling clustered data is Bayesian covariance structure modeling (BCSM) in which the dependence structure is directly modeled through a structured covariance matrix. BCSM have been developed for various applications and complex dependence structures (Fox et al., 2017, Klotzke and Fox, 2019a, 2019b; Mulder and Fox, 2019). This presentation gives an overview of BCSM and discusses several applications/new developments: (1) BCSM for measurement invariance testing (Fox et al., 2020); (2) BCSM for identifying negative within-cluster correlation and personalized (treatment) effects in counseling; and (3) BCSM for interval-censored, clustered, event-time data from a three-armed randomized clinical trial investigating coronary intervention. This talk discusses prior specification, the multiple-hypothesis-testing problem, and computational demands.


Bio: Dr. Jean-Paul Fox is a well-established researcher in the area of Bayesian response modeling. His early work concerns complex psychometric models, for which he received, in 2004 and 2007, a personal grant (an innovational research incentives scheme) from the Netherlands Organisation for Scientific Research (NWO). In 2010, he published a monograph entitled, “Bayesian Item Response Modeling,” covering research in statistics and psychometrics. He has published about 70 refereed journal articles. His more-recent work is about Bayesian covariance structure modeling to improve the modeling of process data in educational measurement and to identify personalized treatment effects. Fox also developed and programmed advanced computational statistical estimation methods to support the application of Bayesian response models for complex data. The software has been published in different repositories, in the Journal of Statistical Software, and on his website ( Several of his publications have described innovative estimation methods and new computational techniques to improve data interpretations. He works at the University of Twente (The Netherlands), where he was appointed Associate Professor in 2007.


Full WEBEX Info for this talk: (Friday, April 23, 2021 2:00 pm)

Meeting number: 120 994 6346

Password: RMMESTAT

Join by video system

You can also dial and enter your meeting number.

Join by phone

+1-415-655-0002 US Toll

Access code: 120 994 6346

Dr. Jennifer Hill, New York University

Friday, 4/30/2021, 12pm

thinkCausal: One Stop Shopping for Answering your Causal Inference Questions

Causal inference is a necessary tool in education research for answering pressing and ever-evolving questions around policy and practice. Increasingly, researchers are using more complicated machine learning algorithms to estimate causal effects. These methods take some of the guesswork out of analyses, decrease the opportunity for “p-hacking,” and are often better suited for more fine-tuned causal inference tasks such as identifying varying treatment effects and generalizing results from one population to another. However, these more sophisticated methods are more difficult to understand and are often only accessible in more technical, less user-friendly software packages. The thinkCausal project is working to address these challenges (and more) by developing a highly scaffolded multi-purpose causal inference software package with the BART predictive algorithm as a foundation. The software will scaffold the researcher through the data analytic process and provide options to access technology-based teaching tools to understand foundational concepts in causal inference and machine learning. This talk will briefly review BART for causal inference and then discuss the challenges and opportunities in building this type of tool. This is work in progress and the goal is to create a conversation about the tool and role of education in data analysis software more broadly.


Bio: Dr. Jennifer Hill develops and evaluates methods that help us answer the causal questions that are vital to policy research and scientific development. In particular she focuses on situations in which it is difficult or impossible to perform traditional randomized experiments, or when even seemingly pristine study designs are complicated by missing data or hierarchically structured data. Most recently Hill has been pursuing two intersecting strands of research. The first focuses on Bayesian nonparametric methods that allow for flexible estimation of causal models and are less time-consuming and more precise than competing methods (e.g. propensity score approaches). The second line of work pursues strategies for exploring the impact of violations of typical causal inference assumptions such as ignorability (all confounders measured) and common support (overlap). Hill has published in a variety of leading journals including Journal of the American Statistical Association, Statistical Science, American Political Science Review, American Journal of Public Health, and Developmental Psychology. Hill earned her PhD in Statistics at Harvard University in 2000 and completed a postdoctoral fellowship in Child and Family Policy at Columbia University’s School of Social Work in 2002. Hill is also the Director of the Center for Practice and Research at the Intersection of Information, Society, and Methodology (PRIISM) and Co-Director of and the Master of Science Program in Applied Statistics for Social Science Research (A3SR).


Full WEBEX Info for this talk: (Friday, April 30, 2021 12:00 pm)

Meeting number: 120 345 7527

Password: RMMESTAT

Join by video system

You can also dial and enter your meeting number.

Join by phone

+1-415-655-0002 US Toll

Access code: 120 345 7527

Dr. David Kaplan, University of Wisconsin-Madison

Friday, 5/21/2021, 12pm

Developments and Extensions in the Quantification of Model Uncertainty: A Bayesian Perspective

Issues of model selection have dominated the theoretical and applied statistical literature for decades. Model selection methods such as ridge regression, the lasso and the elastic net have replaced ad hoc methods such as stepwise regression as a means of model selection. In the end, however, these methods lead to a single final model that is often taken to be the model considered ahead of time, thus ignoring the uncertainty inherent in the search for a final model. One method that has enjoyed a long history of theoretical developments and substantive applications, and that accounts directly for uncertainty in model selection, is Bayesian model averaging (BMA). BMA addresses the problem of model selection by not selecting a final model, but rather by averaging over a space of possible models that could have generated the data. The purpose of this paper is to provide a detailed and up-to-date review of BMA with a focus on its foundations in Bayesian decision theory and Bayesian predictive modeling. We consider the selection of parameter and model priors as well as methods for evaluating predictions based on BMA. We also consider important assumptions regarding BMA and extensions of model averaging methods to address these assumptions, particularly the method of Bayesian stacking. Extensions to problems of missing data and probabilistic forecasting in large-scale educational assessments are discussed.


Bio: Dr. David Kaplan is the Patricia Busk Professor of Quantitative Methods in the Department of Educational Psychology at the University of Wisconsin – Madison. His research focuses on the development of Bayesian statistical methods for education research, with applications to large-scale cross-sectional and longitudinal survey designs. Dr. Kaplan is an elected member of the National Academy of Education and serves as the chair of its Research Advisory Committee; a recipient of the Samuel J. Messick Distinguished Scientific Contributions Award from the American Psychological Association (Division 5); a past-President of the Society for Multivariate Experimental Psychology; a fellow of the American Psychological Association (Division 5); a recipient of the Alexander Von Humboldt Research Award; an Honorary Research Fellow in the Department of Education at the University of Oxford; and a fellow of the Leibniz Institute for Educational Research and Information and the Leibniz Institute for Educational Trajectories. He was also a Jeanne Griffith Fellow at the National Center for Education Statistics. He received his Ph.D. in Education from UCLA in 1987.


Full WEBEX Info for this talk: (Friday, May 21, 2021 12:00 pm)

Meeting number: 120 631 0724

Password: RMMESTAT

Join by video system

You can also dial and enter your meeting number.

Join by phone

+1-415-655-0002 US Toll

Access code: 120 631 0724