Wang Jichuan, Carpenter James R, Kepler Michael A
Center for Interventions, Treatment & Addictions Research, Wright State University School of Medicine, Dayton, OH, USA.
Comput Methods Programs Biomed. 2006 May;82(2):130-43. doi: 10.1016/j.cmpb.2006.02.006. Epub 2006 Mar 29.
In multilevel modeling, researchers often encounter data with a relatively small number of units at the higher levels. As a result, of this and/or non-normality of the residuals, model parameter estimates, particularly the variance components and standard errors of parameter estimates at the group level, may be biased, thus the corresponding statistical inferences may not be trustworthy. This problem can be addressed by using bootstrap methods to estimate the standard errors of the parameter estimates for significance testing. This study illustrates how to use statistical analysis system (SAS) to conduct nonparametric residual bootstrap multilevel modeling. Specific SAS programs for such modeling are provided.
在多水平建模中，研究人员经常遇到高层级单元数量相对较少的数据。因此，由于这一点和/或残差的非正态性，模型参数估计，尤其是组水平上参数估计的方差分量和标准误差，可能会产生偏差，从而相应的统计推断可能不可靠。这个问题可以通过使用自助法来估计用于显著性检验的参数估计的标准误差来解决。本研究说明了如何使用统计分析系统（SAS）进行非参数残差自助多水平建模。提供了用于此类建模的具体SAS程序。
