Practical Significance of Effects from Growth Modeling of Alcohol Use Data

Based on Research Conducted at OSLC

The major goals for this project were to (1) formulate equations for new effect sizes for GMA and GMM hypothesis tests, (2) conduct a secondary analysis study of alcohol use data from both the current and previous National Longitudinal Surveys of Youth, (3) illustrate the use of new statistics in meta-analysis and (4) publish tutorial […]

Project Overview

The major goals for this project were to (1) formulate equations for new effect sizes for GMA and GMM hypothesis tests, (2) conduct a secondary analysis study of alcohol use data from both the current and previous National Longitudinal Surveys of Youth, (3) illustrate the use of new statistics in meta-analysis and (4) publish tutorial articles to widely disseminate derived equations for communicating the practical significance of findings from GMA and GMM.

Year Project Began: 2017
Funder: National Institute of Mental Health

Principal Investigator

Alan Feingold, Ph.D.

Research Scientist
Oregon Social Learning Center

Active Research Projects

Primary Research and Clinical Interests

Dr. Feingold came to OSLC after having worked for nearly a decade as a biostatistician in the Division of Substance Abuse at the Department of Psychiatry at the Yale University School of Medicine. His main current interest at OSLC is in developing and validating new methods of effect sizes (and confidence intervals for these new effect sizes) to convey the practical significance of findings from latent growth, multilevel, and mixture modeling of continuous and categorical outcomes in the alcohol field. Developing and validating effect sizes for the indirect effects from mediation analysis with different kinds of variable distributions is a key aim of his NIAAA-funded work.