论文部分内容阅读
[目的]计算二分类计数资料整群随机化设计的组内相关系数及样本量。[方法]编写SAS宏程序,采用蒙特卡罗模拟方法,估计整群随机化设计的组内相关系数,并根据组内相关系数对样本量进行再计算。[结果]该宏程序可以方便地计算组内相关系数及样本量,并且通过利用SAS软件的ODS功能实现到word的输出。[结论]当组内相关系数未知而无法计算样本量时,该宏程序是可利用的方便工具。
[Objective] To calculate the intra-group correlation coefficient and sample size of two groups of categorical data. [Methods] SAS macro programs were prepared. The Monte Carlo simulation method was used to estimate the intragroup correlation coefficients of the randomized design of the cluster. The sample size was recalculated based on the correlation coefficients in the cluster. [Result] This macro program can easily calculate the correlation coefficient and the sample size in the group, and realize the output of word by utilizing the ODS function of SAS software. [Conclusion] This macro program is a convenient and useful tool when the intra-group correlation coefficient is unknown and the sample size can not be calculated.