Greenhouse-Geisser；统计结果报告；效应力大小介绍

最新推荐文章于 2023-11-01 14:33:08 发布

weixin_30856965

最新推荐文章于 2023-11-01 14:33:08 发布

阅读量3.5k

点赞数 2

CC 4.0 BY-SA版权

原文链接：http://www.cnblogs.com/suho/p/8849698.html

本文介绍了在统计分析中，当球形检验被违反时，如何使用Greenhouse-Geisser和Huynd-Feldt校正来调整F分布的自由度，以获得更准确的显著性值。Greenhouse-Geisser校正通常较为保守，而Huynd-Feldt校正可能过于自由。当ε接近1时，推荐使用Greenhouse-Geisser校正，若ε大于0.75，两种校正均可使用。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

转自博客 http://blog.sina.com.cn/s/blog_4d25466d0101p47z.html

===============================================================================
Greenhouse-Geisser 一般在ANOVA的统计分析常用，在结果报告中我很困惑其df的报告。今天特意把这个问题弄个明白：
自由度是否报告校正后的，让我很困恼，有网友说：即便校正也不需要报告校正后自由度，只报告原来非校正的。或者看看文献里如何报告的。有位发表过脑电文章的在国内科研单位工作的网友说：是必须报告校正后的，也一般报告四舍五入的整数值即可。
如下是源自：https://statistics.laerd.com/statistical-guides/sphericity-statistical-guide-2.php

This is to counteract the fact that when the assumption of sphericity is violated, there is an increase in Type I errors due to the critical values in the F-table being too small. These corrections attempt to correct this bias.---校正的目的
epsilon (referred to as

）具体是指：如下的红线框旁边的（p大于0.05，所以这个是没有违反球形检验）：

An epsilon of 1 (i.e., ε = 1) indicates that the condition of sphericity is exactly met. The further epsilon decreases below 1 (i.e., ε < 1), the greater the violation of sphericity. Therefore, you can think of epsilon as a statistic that describes the degree to which sphericity has been violated.

当ε = 1时，说明这个值就是满足球形检验；但是当这个值越是小于1时，则越不满足违反了球形检验。

Greenhouse-Geisser Correction

Greenhouse-Geisser Correction为了校正F-分布的自由度进行估计epsilon;比如在违反了球形检验，就可以使用该检验。自由度也要相应的变化：

The Greenhouse-Geisser procedure estimates epsilon (referred to as ) in order to correct the degrees of freedom of the F-distribution as has been mentioned previously, and shown below: