R-使用ddply从方差分析中提取p值

Question

I am stuck on a bit of analysis using the ddply function.

The dataframe ( long.format ) structure is as follows:

'data.frame':   2058 obs. of  3 variables:
 $ tertile : Factor w/ 3 levels "1","2","3": 2 1 3 2 2 3 1 1 1 1 ...
 $ variable: Factor w/ 21 levels "age","ht","wt",..: 1 1 1 1 1 1 1 1 1 1 ...
 $ value   : num  57 57 65 58 59 56 63 63 68 58 ...

I have used ddply to perform anova of $variable across $tertile using $value as follows:

aovfun <- function(x)  aov(tertile~value,x)
y <- ddply(long.format, .(variable), aovfun)

The subsequent database gives me the intercept and slope but is there way i can extract individual P values as well?

I have attached a code for an example data.frame as below:

tertile = rep(1L:3L,9)

age=rnorm(9,60,5)

ht=rnorm(9,157,10)

wt=rnorm(9,70,5)

df <- data.frame(tertile,age,ht,wt)

long.format <- melt(df,id=c("tertile"))

Answer 1

Try:

 aovfun <- function(x) aov(value~tertile,x)
 ddply(long.format, .(variable), 
       function(x) summary.aov(aovfun(x))[[1]][["Pr(>F)"]])[,-3]

 #  variable         V1
 #1      age 0.06676654
 #2       ht 0.23854030
 #3       wt 0.88039549