metafor-如何提取F统计量
只是一个简短的问题：
我正在使用metafor软件包进行荟萃分析，并且我正在为3级模型使用rma.mv函数。如何为主持人分析提取F统计量？
提前致谢！
Just a short question:
I am doing a meta-analysis using the metafor package and I am using the rma.mv function for a 3-level model. How do I extract the f-statistic for the moderator analysis?
Thanks in advance!
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
找出这个问题答案的一些策略：
查看函数的文档，特别是描述函数返回内容的
Value部分。对于rma.mv()，请参见此处：https://wviechtb.github.io/metafor /reference/rma.mv.html#value-1
您所询问的 F 统计量可能是来自综合检验的统计量。浏览该部分，我们发现
QM - 主持人综合测试的测试统计数据。或者，我们可以直接检查函数返回的对象。
str()函数对此非常有用，因此只需执行str(model)即可。该函数返回了很多内容（并非所有内容都记录在上述Value部分中），并且您本质上只是在查找似乎包含您所在值的元素寻找。您可以更进一步，使用unclass(model)，它从对象中剥离类并按原样打印列表。但同样的想法 - 只需查看包含 F 统计量的元素返回的内容即可。经过一番搜索，你会再次发现它包含在QM元素中。因此，
model$QM将为您提供 F 统计量。Some strategies for figuring out the answer to this:
Look at the documentation for the function and in particular, the
Valuesection that describes what the function returns. Forrma.mv(), see here:https://wviechtb.github.io/metafor/reference/rma.mv.html#value-1
The F-statistic you are asking about is presumably the one from the omnibus test. Looking through the section, we find
QM - test statistic of the omnibus test of moderators.Alternatively, one can examine the object that is returned by the function directly. The
str()function is quite useful for this, so just dostr(model). There is a lot of stuff that is returned by the function (not all of which is documented in theValuesection described above) and you are essentially just looking for the element that appears to contain the value you are looking for. You can go even further and useunclass(model), which strips the class from the object and just prints the list as is. Same idea though - just look through what is returned for the element that contains the F-statistic. After some searching, you will find again that this is contained in elementQM.So,
model$QMwill give you the F-statistic.