问题
I run the following R commands to make a Dunnett's test and get the summary. How can I access each row of the Linear Hypotheses below, which is part of summary output? Basically I do not know the structure of the summary. I attempted using names() but it seems not working as I do not see any named attribute to give that.
library("multcomp")
Group <- factor(c("A","A","B","B","B","C","C","C","D","D","D","E","E","F","F","F"))
Value <- c(5,5.09901951359278,4.69041575982343,4.58257569495584,4.79583152331272,5,5.09901951359278,4.24264068711928,5.09901951359278,5.19615242270663,4.58257569495584,6.16441400296898,6.85565460040104,7.68114574786861,7.07106781186548,6.48074069840786)
data <- data.frame(Group, Value)
fit <- aov(Value ~ Group, data)
set.seed(20140123)
Dunnet <- glht(fit, linfct=mcp(Group="Dunnett"))
summary(Dunnet)
Simultaneous Tests for General Linear Hypotheses
Multiple Comparisons of Means: Dunnett Contrasts
Fit: aov(formula = Value ~ Group, data = data)
Linear Hypotheses:
Estimate Std. Error t value Pr(>|t|)
B - A == 0 -0.35990 0.37009 -0.972 0.76536
C - A == 0 -0.26896 0.37009 -0.727 0.90012
D - A == 0 -0.09026 0.37009 -0.244 0.99895
E - A == 0 1.46052 0.40541 3.603 0.01794 *
F - A == 0 2.02814 0.37009 5.480 0.00112 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Adjusted p values reported -- single-step method)
回答1:
@MatthewLundberg started to say it, though I didn't find all of the values you wanted with str(Dunnet), so I reverted to str(summary(Dunnet)) and found it:
str(summary(Dunnet))
## List of 10
## $ model :List of 13
## ...
## $ linfct : num [1:5, 1:6] 0 0 0 0 0 1 0 0 0 0 ...
## ...
## $ rhs : num [1:5] 0 0 0 0 0
## $ coef : Named num [1:6] 5.0495 -0.3599 -0.269 -0.0903 1.4605 ...
## ...
## $ vcov : num [1:6, 1:6] 0.0822 -0.0822 -0.0822 -0.0822 -0.0822 ...
## ...
## $ df : int 10
## $ alternative: chr "two.sided"
## $ type : chr "Dunnett"
## $ focus : chr "Group"
## $ test :List of 7
## ..$ pfunction :function (type = c("univariate", "adjusted", p.adjust.methods), ...)
## ..$ qfunction :function (conf.level, adjusted = TRUE, ...)
## ..$ coefficients: Named num [1:5] -0.3599 -0.269 -0.0903 1.4605 2.0281
## .. ..- attr(*, "names")= chr [1:5] "B - A" "C - A" "D - A" "E - A" ...
## ..$ sigma : Named num [1:5] 0.37 0.37 0.37 0.405 0.37
## .. ..- attr(*, "names")= chr [1:5] "B - A" "C - A" "D - A" "E - A" ...
## ..$ tstat : Named num [1:5] -0.972 -0.727 -0.244 3.603 5.48
## .. ..- attr(*, "names")= chr [1:5] "B - A" "C - A" "D - A" "E - A" ...
## ..$ pvalues : atomic [1:5] 0.7655 0.9001 0.9989 0.0176 0.0011
## .. ..- attr(*, "error")= num 0.000755
## ..$ type : chr "single-step"
## ..- attr(*, "class")= chr "mtest"
## - attr(*, "class")= chr [1:2] "summary.glht" "glht"
summary(Dunnet)$test[c('coefficients', 'sigma', 'tstat', 'pvalues')]
## $coefficients
## B - A C - A D - A E - A F - A
## -0.35990210 -0.26895636 -0.09026055 1.46052454 2.02814166
##
## $sigma
## B - A C - A D - A E - A F - A
## 0.3700867 0.3700867 0.3700867 0.4054096 0.3700867
##
## $tstat
## B - A C - A D - A E - A F - A
## -0.9724806 -0.7267388 -0.2438903 3.6025896 5.4801802
##
## $pvalues
## [1] 0.765312271 0.900150707 0.998946706 0.017585149 0.001147749
## attr(,"error")
## [1] 0.0007157397
回答2:
The print method for summary.glht objects displays the desired items in a manner that obscures the underlying structure. After you look at the structure of the summary object with str you still need to extract the values.
The $test element of the summary object has several sub-items, each with the same length, so they can be assembled into a dataframe:
> as.data.frame(summary(Dunnet)$test[ c("coefficients", "sigma", "tstat", "pvalues")] )
coefficients sigma tstat pvalues
B - A -0.35990210 0.3700867 -0.9724806 0.7654036572
C - A -0.26895636 0.3700867 -0.7267388 0.9001056962
D - A -0.09026055 0.3700867 -0.2438903 0.9989441213
E - A 1.46052454 0.4054096 3.6025896 0.0172171050
F - A 2.02814166 0.3700867 5.4801802 0.0009869786
There are two items in the $test item that are not going to be of equal length (since they are functions), so they should not be included in the selections with a vector of names.
来源：https://stackoverflow.com/questions/25225480/access-or-parse-elements-in-summary-in-r