如何执行顺序混合方差分析？

问题描述

我想执行类似混合方差分析（即，受试者之间的因素和受试者内部的因素）的操作，但是对顺序数据执行的操作不是连续的。我试图使用 ordinal 包，特别是clmm（）函数，因为这似乎可以处理混合设计，但是我得到了一些奇怪的结果。这是我正在尝试的：

install.packages("ordinal")
library(ordinal)

#Creating variables and dataframe
#This is made up data since I'm only trying to make it work
#This is the between-subjects factor
Group<-factor(c(rep("Exp",20),rep("Ctr",rep("Exp",20)),levels = c("Exp","Ctr"))
#This is the within-subjects factor
Condition<-factor(c(rep("Con1",40),rep("Con2",rep("Con3",40)),levels = c("Con1","Con2","Con3"))
id<-factor(c(rep(c("01","02","03","04","05","06","07","08","09","10","11","12","13","14","15","16","17","18","19","20","21","22","23","24","25","26","27","28","29","30","31","32","33","34","35","36","37","38","39","40"),3)))
Outcome<-factor(sample(1:3,120,replace = T),levels = c(1,2,3),ordered = T)
Dataset1<-data.frame(id,Group,Condition,Outcome)

mod1<-clmm(Outcome ~ Group * Condition + (1|id),link = "logit",data = Dataset1)
summary(mod1)

但是，结果中包含一条警告

总结.clmm（mod1）： 参数的方差-协方差矩阵未定义

此外，标准误差，z和p值均为NA。如果我尝试在此模型上运行anova（），R会显示下一条消息：

vcov.clm（object，method =“ Cholesky”）中的错误：无法计算vcov： 粗麻布不是正定的

那我做错了什么？也许公式应该用另一种方式写？还是您知道另一种适合您的分析方法？非常感谢您的指导。

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）