Applied Propensity Score Analysis I & II
This two-part seminar series addresses the fundamentals of conducting propensity score analysis for health services research. The series features Dr. Michael Oakes from the University of Minnesota and Dr. Elizabeth Stuart, from Johns Hopkins University, and addresses the basics of understanding propensity score analysis and its use for health research; key analytic strategies conducting propensity score analysis, including matching, weighting, and sub-classification in Stata and R; and practical considerations and common complications when using propensity scores.
Registration can be completed online by clicking on the purchase buttons below, or download and fax the purchase form for the on-demand seminar to 202.292.6800, Attn: Online Learning. Individual registration is required for each participant. For more information please contact Courtney Segal at courtney.segal@academyhealth.org.
Applied Propensity Score Analysis I: Rationale and Basic Techniques
Participants of the live webinar should enter their AcademyHealth username and password used for registration to access the on-demand video. If you need assistance with your AcademyHealth username or password please contact Courtney Segal at courtney.segal@academyhealth.org.
Course Level: 201 (Intermediate)
Faculty: J. Michael Oakes, PhD, McKnight Presidential Fellow, Associate Professor, University of Minnesota
Format: On-demand, streaming presentation with voice-over
Registration Price: $25 student members, $50 members, $100 non-members
Learning objectives for part I of the series include:
- Understanding propensity score analysis is and why it might be conducted.
- Appreciating the differences between a propensity score analysis and a common regression-based analysis.
- Understanding the basic procedures involved in a propensity score analysis.
- Appreciating the major shortcomings of propensity score analyses.
Applied Propensity Score Analysis II: Practical Considerations and Lessons for Use
Participants of the live webinar should enter their AcademyHealth username and password used for registration to access the on-demand video. If you need assistance with your AcademyHealth username or password please contact Courtney Segal at courtney.segal@academyhealth.org.
Course Level: 201 (Intermediate)
Faculty: Elizabeth A. Stuart, PhD, Assistant Professor, Johns Hopkins Bloomberg School of Public Health
Format: On-demand, streaming presentation with voice-over
Registration Price: $25 student members, $50 members, $100 non-members
Learning objectives for part II of the series include:
- Learning how to implement propensity score matching, weighting, and sub-classification in R
- The ability to diagnose whether or not a propensity score method "worked" at balancing the covariates.
- Understanding some strategies to deal with some of the common complications in using propensity scores
Faculty Bios:
J. Michael Oakes, PhD, is a McKnight Presidential Fellow and associate professor in the Division of Epidemiology, University of Minnesota. He is also a fellow of the MN Population Research Center and an adjunct professor of sociology. His professional interests center on quantitative methodology, social epidemiology, and research ethics. He is an active researcher and frequent principal investigator on studies addressing a vast array of methodological, health, and ethical topics. Dr. Oakes has authored over 70 papers exploring problems at the intersection of social and medical sciences; his first text entitled Methods in Social Epidemiology was released in 2006. He is widely known for his seminal work on how social stratification and social interaction complicates inferences drawn from many common research designs and statistical methods in public health research. Among widely cited papers are his 2003 paper on measuring SES and his 2004 and 2007 papers on identification of neighborhood effects. His book chapter on the theory behind and practice of propensity score matching methods in social epidemiology is highly acclaimed. Dr. Oakes teaches several graduate-level courses in statistical methods and social epidemiology, including group-randomized trials and advanced epidemiologic methods. He serves the editorial board of Evaluation Review and is senior scientific advisor for the Robert Wood Johnson Foundation's Healthy Eating Research initiative.
Elizabeth A. Stuart, PhD, is an assistant professor in the Department of Mental Health and the Department of Biostatistics at Johns Hopkins Bloomberg School of Public Health. She received her PhD in statistics in 2004 from Harvard University, working under the direction of Donald Rubin. Dr. Stuart has extensive experience in methods for estimating causal effects, particularly as applied to mental health, prevention, and education. Her publications include applied and theoretical papers on the use of matching methods such as propensity scores and she has written software for implementing matching methods in R. She also has extensive experience with designing and analyzing randomized experiments, multilevel modeling, and bayesian methodology. Prior to her time at Hopkins she was a researcher at Mathematica Policy Research, Inc. and she has served as a consultant on propensity score methods for the RAND Corporation, Genzyme, Inc., the Urban Institute, and the American Institutes of Research. She also has extensive teaching experience, including semester-long and one-day short courses on the estimation of causal effects.
System Requirements
The latest version of Adobe Flash Player is required to view on-demand seminars. Download Adobe Flash Player for free.
Please check back for updates on new online training opportunities. For more information or to sign up for notification about new opportunities please contact Courtney Segal at 202.292.6700 or send a message to hsrmethods@academyhealth.org.
