中文摘要：

在许多实际问题中，检验观察数据是否出现异方差性是一个相当感兴趣的问题.该文研究了半参数随机效应模型的异方差检验问题．基于Lin（1997）的方法，得到了检验方差成分都为零的Score检验统计量．通过随机模拟和实际数值例子，论证了方法的有效性．利用现有的统计软件，容易实现该文所提出的检验方法．

英文摘要：

It is of considerable interest in testing for heteroscedasticity in many practical studies. In this paper, the authors discuss this type of problem in framework of semiparametric mixed models. A global score test is proposed for the null hypothesis that all the variance components are zero. The test is based on the approaches of Lin （1997）, the authors examine its performance in a simulation study, illustrate the test methods by using the progesterone data. The test can be easily implemented by using existing statistical softwares.