Workshop on Advanced Research Methods (WARM)
Sharing innovations and innovative uses of advanced research methods, Thursdays 8:30-9:30 PM (US ET)
Organizing Committee: Weihua An, Shawn Bauldry, Yan Long, Ian Lundberg, Xi Song, Jun Xu
2023 Schedule
(Register for selected meetings. Limited spots, first come first serve.)
1. Causal Network Analysis
Time: February 23, 8:30-9:30 PM (US ET)
Speaker: Weihua An, Emory University
Abstract: Fueled by recent advances in statistical modeling and the rapid growth of network data, social network analysis has become increasingly popular in sociology and related disciplines. However, a significant amount of work in the field has been descriptive and correlational, which prevents the findings from being more rigorously translated into practices and policies. This talk provides a review of the popular models and methods for network analysis, with a focus on causal inference threats (such as measurement error, missing data, network endogeneity, contextual confounding, simultaneity, and collinearity), potential solutions (such as instrumental variables, specialized experiments, and leveraging longitudinal data), and future directions for causal network analysis. [Related Paper] [Futher Readings]
2. Bayesian Analysis: An Overview
Time: March 23, 8:30-9:30 PM (US ET)
Speaker: Jun Xu, Ball State University
Abstract: Once portrayed as a heretical paradigm and subjective doctrine, Bayesian inference has emerged from this abject oblivion to a tidal wave to sweep through the world of statistics and data science. This talk begins with the origin of Bayesian statistics, the Bayes theorem, and recounts how and (possibly) why this framework was created. Formerly called the inverse probability approach, and probably more appropriately—Laplacian statistics—Bayesian statistics has undergone the nadir and zenith of its practice, due in part to its computational inconvenience and subjective assignment of priors. With the computational breakthroughs, especially those in the 1980s and early 1990s, several seemingly unrelated dots were connected to create the Markov chain Monte Carlo (MCMC) methods. This has completely changed the landscape in the field and revolutionized the estimation methods for Bayesian statistics. Unlike the classical frequentist statistics with the null hypothesis significance testing (NHST), Bayesian statistics usually uses Bayes factors, probabilities (not the confusing and problematic p-values), and credible intervals (not confidence intervals) to make inferences. Along with prior information integrated into the current iteration of estimation, the Bayesian approach dovetails well with how information is processed and updated epistemologically. This talk is based on the introductory sections of this recently published book. [Resources]
3. Structural Equation Models: Applications and Frontiers
Time: April 20, 8:30-9:30 PM (US ET)
Speaker: Shawn Bauldry, Purdue University
Abstract: Structural Equation Models (SEMs) remain an important component of the social science analytic toolkit. SEMs permit the specification of models with multiple endogenous variables and the presence of latent variables that can address measurement error. This talk will provide an introduction to the general SEM and the workflow of an SEM-based analysis. In addition, we will compare SEMs with directed acyclic graphs (DAGs), a contemporary framework for causal analysis with many similarities and a few important differences. Finally, we will explore a couple of prominent extensions to SEMs, latent growth models, and latent class analysis, that are currently widely used in the social and behavioral sciences. [Presentation Slides] [A Related Review Paper]
4. Sequence Analysis: Its Past, Present, and Future
Time: September 21, 8:30-9:30 PM (US ET)
Speaker: Tim Liao, State University of New York at Stony Brook
Abstract: The application of sequence analysis (SA) in the social sciences, especially in life course research, has mushroomed in the last decade and a half. Using a life course analogy, I examine in this talk the birth of SA in the social sciences and its childhood (the first wave), its adolescence and young adulthood (the second wave), and its future mature adulthood in the paper. The talk provides a summary of (1) the important SA research and the historical contexts in which SA was developed by Andrew Abbott, (2) a brief review of the many methodological developments in visualization, complexity measures, dissimilarity measures, group analysis of dissimilarities, cluster analysis of dissimilarities, multidomain/multichannel SA, dyadic/polyadic SA, Markov chain SA, sequence life course analysis, sequence network analysis, SA in other social science research, and software for SA, and (3) reflections on some future directions of SA including the methods currently being developed, and some remaining challenges facing SA for which we do not yet have any solutions. The talk builds on this recently published paper in Social Science Research.
5. Machine Learning for Social Science
Time: October 19, 8:30-9:30 PM (US ET)
Speaker: Ian Lundberg, Cornell University
Abstract: This talk is based on a paper with Jennie Brand and Nanum Jeon. Computational power and big data have created new opportunities to explore and understand the social world. A special synergy is possible when social scientists combine human attention to certain aspects of the problem with the power of algorithms to automate other aspects of the problem. We review selected exemplary applications where machine learning amplifies researcher coding, summarizes complex data, relaxes statistical assumptions, and targets researcher attention to further social science research. We aim to reduce perceived barriers to machine learning by summarizing several fundamental building blocks and their grounding in classical statistics. We present a few guiding principles and promising approaches where we see particular potential for machine learning to transform social science inquiry. We conclude that machine learning tools are increasingly accessible, worthy of attention, and ready to yield new discoveries for social research. [Presentation Slides]
6. Text Analysis for Social Science
Time: November 17, 8:30-9:30 AM (US ET). Note this is on Friday morning.
Speaker: Ana Macanovic, Utrecht University
Abstract: The emergence of big data and computational tools has introduced new possibilities for using large-scale textual sources in social science research. We discuss five computational text analysis methods that can help researchers analyze large quantities of textual data and discuss exemplary applications in recently published research. First, we show how dictionary methods can assist the quantification of concepts of interest in texts; then, we summarize the potential of using semantic text analysis to extract information on social actors, social actions, and relationships between them. We move on to explore how unsupervised machine learning clustering methods assist inductive exploration of underlying meanings and concepts present in texts and how supervised machine learning classification methods support replication of manual coding onto new data. Finally, we discuss how powerful language models can help us map complex meanings, explore the evolution of meanings over time, and follow the emergence of new concepts in texts. We conclude by emphasizing the important implications of using large datasets and computational methods to infer complex meaning from texts in social sciences. This talk builds on this recently published paper. [Presentation Slides]
Recording: https://youtu.be/oiRRtv_NDL4