R - How to recode multiple columns

Question

I am trying to change the 6s to NAs across multiple columns. I have tried using the mutate_at command in dplyr, but can't seem to make it work. Any ideas?

library(dplyr)
ID <- c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10) #Create vector of IDs for ID column.
Score1 <- c(1, 2, 3, 2, 5, 6, 6, 2, 5, 4) #Create vector of scores for Score1 column.
Score2 <- c(2, 2, 3, 6, 5, 6, 6, 2, 3, 4) #Create vector of scores for Score2 column.
Score3 <- c(3, 2, 3, 4, 5, 5, 6, 2, 6, 4) #Create vector of scores for Score3 column.
df <- data.frame(ID, Score1, Score2, Score3) #Combine columns into a data frame.
VectorOfNames <- as.vector(c("Score1", "Score2", "Score3")) #Create a vector of column names.
df <- mutate_at(df, VectorOfNames, 6=NA) #Within the data frame, apply the function (6=NA) to the columns specified in VectorOfNames.

score 1 · Accepted Answer · answered Apr 06 '20 at 05:02

dplyr has the na_if() function for precisely this task. You were almost there with your code and can use:

mutate_at(df, VectorOfNames, ~na_if(.x, 6))

   ID Score1 Score2 Score3
1   1      1      2      3
2   2      2      2      2
3   3      3      3      3
4   4      2     NA      4
5   5      5      5      5
6   6     NA     NA      5
7   7     NA     NA     NA
8   8      2      2      2
9   9      5      3     NA
10 10      4      4      4

score 0 · Answer 2 · answered Apr 06 '20 at 04:53

You could use :

library(dplyr)
df %>%mutate_at(VectorOfNames, ~replace(., . == 6, NA))
#OR
#df %>%mutate_at(VectorOfNames, ~ifelse(. == 6, NA, .))


#   ID Score1 Score2 Score3
#1   1      1      2      3
#2   2      2      2      2
#3   3      3      3      3
#4   4      2     NA      4
#5   5      5      5      5
#6   6     NA     NA      5
#7   7     NA     NA     NA
#8   8      2      2      2
#9   9      5      3     NA
#10 10      4      4      4

Or in base R :

df[VectorOfNames][df[VectorOfNames] == 6] <- NA

R - How to recode multiple columns

2 Answers2