Hosted by the GPH Office of Research & Department of Biostatistics
Join Dr. Chuck Huber, Director of Statistical Outreach at StataCorp, who will be giving three consecutive lectures on Structural Equation Modeling, Latent Class Analysis, and an Introduction to Mediation Analysis, with applications in Stata. More details about each lecture can be found below.
In-person spots are open only to current NYU students, staff, and faculty. The general public will be able to participate virtually.
Lecture 1: Structural Equation Modeling (10:00-11:30am)
Dr. Huber introduces the concepts and jargon of structural equation modeling (SEM) including path diagrams, latent variables, endogenous and exogenous variables, and goodness of fit. He describes the similarities and differences between Stata's -sem- and -gsem- commands. Then, he demonstrates how to fit many familiar models such as linear regression, multivariate regression, logistic regression, confirmatory factor analysis, and multilevel models using -sem- and -gsem-. He wraps up by demonstrating how to fit structural equation models that contain both structural and measurement components.
Lecture 2: Latent Class Analysis Using Stata (11:30am-12:30pm)
Latent variables are a useful tool for modeling hypothetical constructs such as intelligence, ability, depression, and anxiety. Models for categorical latent variables are often called Latent Class Analysis (LCA) or Latent Profile Analysis (LPA). The levels of a categorical latent variable represent groups in the population and are called classes. We are interested in identifying and understanding these classes. LCA is characterized by discrete response variables while LPA is characterized by continuous response variables. This presentation will provide a brief introduction to LCA/LPA models and how to implement them using Stata.
Lecture 3: Introduction to Mediation Analysis (12:30-1:30pm)
Mediation analysis is a statistical technique that is used to parse the direct, indirect, and total effect an independent variable (X) on a dependent variable (Y) in the presence of a mediating variable (M). For example, age (X) is associated with higher systolic blood pressure (Y). Age is also associated with higher weight (M) and increased weight is associated with higher systolic blood pressure (SBP). Age has a direct effect on SBP as well as an indirect effect on SBP through its effect on weight. This talk will introduce the concepts of mediation analysis and demonstrate how to conduct mediation analysis using regression, structural equation modeling (SEM), and how to estimate bootstrap standard errors for mediation models with SEM.
About the Speaker:
Dr. Chuck Huber is Director of Statistical Outreach at StataCorp and Adjunct Associate Professor of Biostatistics at the Texas A&M School of Public Health and at the New York University School of Global Public Health. In addition to working with Stata's team of software developers, he produces instructional videos for the Stata Youtube channel, writes blog entries, develops online NetCourses and gives talks about Stata at conferences and universities. Most of his current work is focused on statistical methods used by behavioral and health scientists. He has published in the areas of neurology, human and animal genetics, alcohol and drug abuse prevention, nutrition and birth defects. Dr. Huber currently teaches survey sampling at NYU and introductory biostatistics at Texas A&M where he previously taught categorical data analysis, survey data analysis, and statistical genetics.