**CO-SPONSORED METHODS WORKSHOP: Machine Learning**

Date/Time
Date(s) - Mar 18, 2021
12:00 pm - 2:00 pm

Categories


NIA Aging Centers’ Collaborative Virtual Methods Workshop Series

The workshop series is co-sponsored by: 
– Center for Aging and Policy Studies, Syracuse/Cornell/U-Albany
– Center for Advancing Sociodemographic and Economic Study of Alzheimer’s Disease, USC/Stanford/UT Austin
– Center on Aging & Population Sciences, UT Austin
– Center for the Demography of Health and Aging, UW-Madison.

This particular workshop is organized by:
Center for Aging and Policy Studies, Syracuse/Cornell/U-Albany

Workshop Title: Uncovering Sociological Effect Heterogeneity using Machine-Learning

Speaker: Jennie E. Brand

Workshop Description: Individuals do not respond uniformly to treatments, such as events or interventions. Sociologists routinely partition samples into subgroups to explore how the effects of treatments vary by selected covariates, such as race and gender, based on theoretical priors. Data-driven discoveries are also routine, yet the analyses by which sociologists typically go about them are often problematic, and seldom move us beyond our biases to explore new meaningful subgroups. In this workshop, I present emerging machine learning methods based on decision trees that allow researchers to explore sources of variation that they may not have previously considered or envisaged. I demonstrate tree-based machine learning, that is, causal trees, to recursively partition the sample to uncover sources of effect heterogeneity. Assessing a central topic in social inequality, college effects on wages, I will compare what we learn from covariate and propensity-score-based partitioning approaches to recursive partitioning based on causal trees. Decision trees, although superseded by forests for estimation, can be used to uncover subpopulations responsive to treatments. Using observational data, I expand on the existing causal tree literature by applying leaf-specific effect estimation strategies to adjust for observed confounding, including inverse propensity weighting, nearest neighbor matching, and doubly robust causal forests. I also assess localized balance metrics and sensitivity analyses to address the possibility of differential imbalance and unobserved confounding. We will demonstrate the R package to perform these methods and data visualization techniques.

Bio: Jennie E. Brand is Professor of Sociology and Statistics. She is also Director of the California Center for Population Research and Co-Director of the Center for Social Statistics (CSS). Prof. Brand studies social stratification and inequality, and its implications for various outcomes that indicate life chances. Her research agenda encompasses three main areas: (1) access to and the impact of higher education; (2) the consequences of disruptive events, such as job displacement; and (3) causal inference and quantitative and computational methods for observational data. She is Chair of the Methodology Section of the American Sociological Association (ASA), Chair-Elect of the Inequality, Poverty, and Mobility Section of ASA, and an elected Board Member of the International Sociological Association (ISA) Research Committee on Social Stratification and Mobility (RC28). She was elected to the Sociological Research Association (SRA), an honor society for excellence in research, in 2019, and received the ASA Methodology Leo Goodman Mid-Career Award in 2016, and honorable mention for the ASA Inequality, Poverty, and Mobility William Julius Wilson Mid-Career Award in 2014. Prof. Brand is a member of the Technical Review Committee for the National Longitudinal Surveys Program at the Bureau of Labor Statistics. She was previously a member of the Board of Overseers of the General Social Survey (GSS). She is Associate Editor of AAAS’s Science Advances, the open access extension of Science magazine.

Registration for this event is now closed.