1

这是我的模型：

mod6<-glmer(NS.test~0+Beach*method+(1|year)+ (1|Beach/Zone_1/index),data=y2,weights=(y2$FC+y2$Nest),family=binomial)

我做了多重比较：

    wht <- glht(mod6, linfct = mcp(Beach = "Tukey", method = "Tukey"))

摘要（wht，测试=调整（“西部荒野”））

结果：

> wht <- glht(mod6, linfct = mcp(Beach = "Tukey", method = "Tukey"))
Warning messages:
1: In mcp2matrix(model, linfct = linfct) :
  covariate interactions found -- default contrast might be inappropriate
2: In mcp2matrix(model, linfct = linfct) :
  covariate interactions found -- default contrast might be inappropriate
    > summary(wht, test = adjusted("Westfall"))

     Simultaneous Tests for General Linear Hypotheses

Multiple Comparisons of Means: Tukey Contrasts


Fit: glmer(formula = NS.test ~ 0 + Beach * method + (1 | year) + (1 | 
Beach/Zone_1/index), data = y2, family = binomial, weights = (y2$FC + 
y2$Nest))

Linear Hypotheses:
                                   Estimate Std. Error z value Pr(>|z|)    
Beach: HO/HA - Hillsboro == 0       0.28208    0.13787   2.046   0.0408 *  
Beach: P/FTL - Hillsboro == 0       0.45910    0.11874   3.866   <0.001 ***
Beach: P/FTL - HO/HA == 0           0.17703    0.06999   2.529   0.0196 *  
method: HTL only - HTL and SB == 0  0.64003    0.13472   4.751   <0.001 ***
method: no clean - HTL and SB == 0  0.36418    0.08201   4.441   <0.001 ***
method: no clean - HTL only == 0   -0.27585    0.10690  -2.581   0.0196 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Adjusted p values reported -- Westfall method)

这是我发现多重比较起作用的唯一方法。如您所见，我有一个错误。知道这意味着什么吗？

如何用上面的字母绘制结果？

是否有另一种方法可以通过另一种方式对交互进行多重比较？

4

0 回答 0