Please read the question completely and take care about that dplyr functions use non-standard evaluation (NSE). That makes use of spread or dcast beside dplyr commands inside a function somewhat problematic. My question is not about converting long to wide Format in normal case but inside a function including dplyr commands!
Suppose the following data frame df:
set.seed(1357)
df <- data.frame(
"id" = seq(1:100),
"grp" = sample(c("A","B","C","D"), 100, replace=TRUE, prob=c(0.25, 0.25, 0.30, 0.20)),
"var1" = sample(c("X","Y"), 100, replace=TRUE, prob=c(0.45, 0.55)),
"var2" = sample(c("X","Y"), 100, replace=TRUE, prob=c(0.15, 0.85)),
"var3" = sample(c("X","Y"), 100, replace=TRUE, prob=c(0.50, 0.50)),
"var4" = sample(c("X","Y"), 100, replace=TRUE, prob=c(0.40, 0.60)),
"var5" = sample(c("X","Y"), 100, replace=TRUE, prob=c(0.10, 0.90)),
"var6" = abs(rnorm(100)))
I want to provide a frequncy table for grp and all ather variable var1 to var5 via lapply. For this purpose I wrote a function named fnct.freq.
library(dplyr)
library(tidyr)
library(reshape2)
fnct.freq <- function(x){
x <- enquo(x)
freqtab <- df %>%
select(grp, !!x) %>%
group_by(grp, !!x) %>%
summarise(n = n()) %>%
mutate(freq = n / sum(n))
freqtab$nfreq <- paste0(freqtab$n, " (", round(freqtab$freq,2), ")")
freqtab <- freqtab[,-c(3,4)]
# ???? here I want to convert long to wide format
# ???? some code like this:
# dcast(freqtab, noquote(x) ~ grp, value.var="nfreq") # dose not work
return(freqtab)
}
fnct.freq(var2)
The providing result is:
> fnct.freq(var2)
# A tibble: 8 x 3
# Groups: grp [4]
grp var2 nfreq
<fct> <fct> <chr>
1 A X 1(0.06)
2 A Y 16(0.94)
3 B X 2(0.07)
4 B Y 26(0.93)
5 C X 3(0.08)
6 C Y 33(0.92)
7 D X 4(0.21)
8 D Y 15(0.79)
as you can see the result is a frequency table in a long format. But I want to have it in the wide format via packages like as rshape2 or tidyr. The result should be
grp X_nfreq Y_nfreq
1 A 1 (0.06) 16 (0.94)
2 B 2 (0.07) 26 (0.93)
3 C 3 (0.08) 33 (0.92)
4 D 4 (0.21) 15 (0.79)
Later I want to apply the function on other variables of data frame
lapply(c("var1","var2","var3","var4","var5"), FUN=fnct.freq)
