The power of the standard test for the presence of heterogeneity in meta-analysis
The power of the standard test for the presence of heterogeneity in meta-analysis
复制标题
DOI：
10.1002/sim.2481
复制
发表时间：
2006-08-15
影响因子：
2
通讯作者：
Jackson, Dan
中科院分区：
文献类型：
--
作者：
Jackson, Dan
有人建议，荟萃分析中异质性存在的标准检验功效较低。尽管已经使用模拟对此进行了研究，但几乎没有直接的分析证据证明该主张的有效性。使用已建立的检验统计量的近似分布，描述了获得检验功效的过程。由此，可以得到一个简单的功率公式。尽管这适用于特殊情况，但该公式更普遍地表明了检验的功效。特别是，对于给定的显着性水平，可以根据研究间变异提供的研究方差的比例来校准功效。对该数量的考虑证实了该测试通常具有较低的功效。建议希望进行标准测试的从业者在执行测试之前使用所提供的想法来调查测试的操作特性。版权所有 (c) 2005 John Wiley & Sons, Ltd.
It has been suggested that the standard test for the presence of heterogeneity in meta-analysis has low power. Although this has been investigated using simulation, there is little direct analytical evidence of the validity of this claim. Using an established approximate distribution for the test statistic, a procedure for obtaining the power of the test is described. From this, a simple formula for the power is obtained. Although this applies to a special case, the formula gives an indication of the power of the test more generally. In particular, for a given significance level, the power can be calibrated in terms of the proportion of the studies' variances that is provided by between-study variation. A consideration of this quantity confirms that the test does, in general, have low power. It is suggested that practitioners, who wish to conduct the standard test, use the ideas provided in order to investigate the operating characteristics of the test prior to performing it. Copyright (c) 2005 John Wiley & Sons, Ltd.