1

I have a dataframe df, which has variables component1_done, component2_done and so on. I want to table each, for example, as:

proportions(table(df$component1_done))
proportions(table(df$component2_done))
proportions(table(df$component3_done))
proportions(table(df$component4_done))
proportions(table(df$component5_done))
proportions(table(df$component6_done))

How can I do the same using a loop? In other words, how can I refer to df$component"i"_done in a loop?

Thanks!

Toshi Furukawa
  • 11
  • 1
  • The answers https://stackoverflow.com/questions/18222286/dynamically-select-data-frame-columns-using-and-a-character-value and https://stackoverflow.com/questions/12389318/dollar-sign-before-a-variable show the standard way to extract values when using a character string e.g. you dont use the `$` notartion – user20650 Dec 31 '21 at 01:56

4 Answers4

1

You can use lapply for that. The result is a list of proportions for all columns.

df <- data.frame(a=c(1,2,2,4,5), b=c(3,4,5,6,6))

lapply(df, function(x) proportions(table(x)))
$a
x
  1   2   4   5
0.2 0.4 0.2 0.2

$b
x
  3   4   5   6
0.2 0.2 0.2 0.4

With dplyr

library(dplyr)

as.list(df %>% summarise(across(everything(), ~ proportions(table(.x)))))
$a
a
  1   2   4   5
0.2 0.4 0.2 0.2

$b
b
  3   4   5   6
0.2 0.2 0.2 0.4
Andre Wildberg
  • 12,344
  • 3
  • 12
  • 29
  • 1
    Thank you so much, Andre! Yours is of course much prettier, but in the meanwhile I have also struggled to use a loop. Our answer was to use a longer format. – Toshi Furukawa Dec 31 '21 at 02:55
0

Thank you so much, Andre! Yours is of course much prettier, but in the meanwhile I have also struggled to use a loop. Our answer was to use a longer format.

for (i in 1:6){
  df0 <- df %>% filter(answerindex==8)
  char <- paste0("component",i,"_done")
  df0 <- df0 %>% select("username",all_of(char))
  df0_longer <-  pivot_longer(df0, col = -c(username), names_to = "component", values_to = "done")
  t <- proportions(table(df0_longer$done))
  print(t)
}

This works but I was very unhappy that this was longer than the original six lines ;)

Toshi Furukawa
  • 11
  • 1
0

If you really want to use a loop you can use

df <- data.frame(a=c(1,2,2,4,5), b=c(3,4,5,6,6))

for (i in seq_along(names(df))){
       print(proportions(table(df[,i])))
    }

Loops don't print unless you say to.
Alternatively you can assign each table to a list and then print the list.

tables <- list()
for (i in seq_along(names(df))){
       tables[[i]] <- proportions(table(df[,i]))
    }
tables
Elin
  • 6,507
  • 3
  • 25
  • 47
0

O, my, I found an even easier one ("apply" functions were good, but using the following "for" is more intuitive to me for the moment. Probably I will simply have to get used to "apply" family functions.)

for (i in 1:6){
  print(paste0("component",i,"_done"))
  print(proportions(table(df[,paste0("component",i,"_done")])))
}

The trick is: you cannot name a variable by writing df$paste0("a","b") but you can name the variable by writing df[,paste0("a","b")]

Toshi Furukawa
  • 1