234662 SAS Macro for Nonparametric Bootstrap Multilevel Modeling

Monday, October 31, 2011: 2:30 PM

Jichuan Wang, PhD , School of Medicine, The George Washington University, Bowie, MD
Peichang Shi , Department of Pharmaceutical Health Services Research, University of Maryland, Baltimore
Avital Cnaan, PhD , School of Medicine and Health Sciences, George Washington University, Washington, DC
Multilevel models have been widely used to analyze hierarchically structured data. However, researchers often encounter a challenging issue with their data; i.e., the number of higher level units (e.g., the second level units) is sometimes small. Both parametric and non-parametric residual bootstrap multilevel modeling can be applied to address the problem in multilevel modeling with small a number of higher units. The authors of the present study have developed a user-friendly SAS macro RBMLM for such modeling. In the proposed presentation, the authors will discuss the theory of residual bootstrap multilevel modeling and the strategy of conducting such modeling. Real data will be used to demonstrate how to apply the SAS macro to conducting non-parametric residual bootstrap multilevel modeling, and the results will be compared with those from the regular multilevel model. The SAS macro can also be used for non-parametric residual bootstrap multilevel modeling when data do not meet the normality assumption.

Learning Areas:
Biostatistics, economics
Public health or related research
Social and behavioral sciences

Learning Objectives:
1) Discuss the problems in multilevel modeling when the number of higher level units is small. 2) Discuss bootstrap residual multilevel modeling, including parametric and non-parametric residual bootstrap. 3) Demonstrate how to use the SAS macro RBMLM.

Keywords: Statistics, Computer-Assisted

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am qualified on the content I am responsible for because I have been working in public health studies for over 20 years. I usually do two presentations in the Statistics Sessions of the APHA annual meetings each year in many of the past years.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.