American Statistical Association
Item response theory (IRT) modeling and factor analysis (FA) are two statistical techniques useful for examining the patterns among a set of symptoms or criteria associated with psychiatric disorders. Item response analysis (originally developed and enhanced within the field of educational testing) has played an important role in identifying psychiatric symptoms along a continuum of severity of disorder and has been used to help inform proposed changes to DSM-5. Factor analysis (originally developed and enhanced in psychology) has also played an important role in identifying psychiatric symptom groups and categorizing symptoms into different sub-dimensions or constructs. I have labeled this seminar a tutorial because the goal is to provide a kind of "how-to" for the audience to learn how they can use these two statistical methods for their own measurement problems. I will present the methods in the context of two examples including the measurement of nicotine use disorder and the measurement of complicated grief, both of which are based on collaborations with colleagues at NYSPI.
Dr. Wall is a Professor of Biostatistics (in Psychiatry) at Columbia University. She has a Ph.D. in statistics from Iowa State University (1998) and was previously on the faculty in the School of Public Health at the University of Minnesota from 1998-2010. Her research has predominately focused on the development and improvement of biostatistical models and methods for research problems where the target variables of interest are latent, specifically variables that may be conceptual in nature or simply not measured directly. She also has made several methodological contributions in the areas of longitudinal and geographical (or spatial) data analysis. Most of Dr. Wall’s collaborative work has involved answering research questions in social/behavioral public health and psychiatry using observational studies based on large multilevel data sources as well as small clinical samples. In spring 2012, Dr. Wall will offer a course entitled “Latent Variable and Structural Equation Modeling for Health Sciences” through the Department of Biostatistics in the Mailman School of Public Health.
|Date:||Tuesday, October 4, 2011|
|Time:||3:30 - 4:30 P.M.|
New York State Psychiatric Institute
1051 Riverside Drive
6th Floor Multipurpose Room (6602)
New York, New York