Introduction to Propensity Score Analysis Using R for Social Work Research and Program Evaluation

Statistical packages such as SAS, SPSS, and Stata are widely used in academic and research institutions. However, due to their high license costs, some social work programs choose Excel statistical toolpak for teaching data analysis despite its statistical limitations.  The statistical software R has emerged as an increasingly popular tool as a free alternative because of its public availability, flexibility and abilities to handle most advanced statistical analyses. Recently, interest in using R has significantly grown among many social work researchers (Auerbach & Zeitlin, 2014; Auerbach, Zeitlin & Mason, 2015; Passolt, Smith & Mienko, 2015; Tran & Wang, 2015). However, information regarding advanced statistical analyses using R remains scarce in social work literature.

Propensity score analysis is an advanced statistical method to reduce or control for potential biases in the estimation of treatment outcomes or dependent variables, especially in non-randomized clinical trials or observational studies. The purpose of this workshop is to teach attendants to conduct propensity score analysis using R. We will first provide an overview of propensity score analysis, and then demonstrate the application of R to estimate propensity score for matching, stratification and adjustment in ordinary least square and logistic regressions. We will also address the strengths and limitations of this method.

Topics    

• Overview of propensity score.
• Application of propensity score in matching
• Application of propensity score in stratification
• Application of propensity score for adjustment in regression
• Strengths and limitations of propensity score analysis in observational studies.

Career Level and Prerequisites

Participants are expected to have knowledge in regression analysis, research methods and design. Experience with the R language is recommended, but not required. At the end of the workshop, participants will have basic understanding of propensity score analysis using R.  Hands-on learning materials will be available at the workshop.

Methods and Approach

This hands-on workshop will be conducted in an interactive manner between the attendants and the presenters. Powerpoint presentation and data analysis demonstration will be used with interpretations and discussions. Handouts will be provided. Attendants are welcome to ask questions during the session.

Recommended References

Austin, P. C. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate behavioral research, 46(3), 399-424.

D’Agostino, R. B. (1998). Tutorial in biostatistics: propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Statistics in Medicine, 17(19), 2265-2281.

Guo, S., & Fraser, M. W. (2014). Propensity score analysis: Statistical methods and applications. Sage Publications.

Susanne Stampf (2014). nonrandom: Stratification and matching by the propensity score. R package version 1.42. Retrieved from: http://CRAN.r-project.org/package=nonrandom