how to recode (and reverse code) variables in columns with dplyr

Question

I am picking up R again after last using it in 2013. I am getting used to using dplyr, but I am running into a problem with a simple task. I have a table that looks like

Participant Q1       Q2      Q3     Q4       Q5
1           agree  neutral   NA    Disagree  Agree
2           neutral agree    NA     NA       NA

My goal

   Participant Q1       Q2      Q3     Q4       Q5
    1           3       2       NA      1       3
    2           2       1       NA     NA       NA

I want to be able to change the categorical value to a numerical value for columns Q1:Q5, but all the examples that I see of using recode for dplyr work for rows and no columns. (I might be missing something in the examples). I then want to be able to pick column Q1 and Q2 and reverse code it.

I am trying to learn to do this in dplyr if possible

Thanks

Answer 1

This is fairly straightforward now thanks to dplyr's recode function. Here's one way to do it:

# Generate a dataframe to match yours

df <- data.frame(
  participant = c(1,2),
  Q1 = c("agree", "neutral"),
  Q2 = c("neutral", "agree"),
  Q3 = c(NA,NA),
  Q4 = c("Disagree", NA),
  Q5 = c("Agree", NA)
)

# Use recode to recode the data

df_recode <- df %>%
  mutate(Q1 = recode(Q1, "agree" = 3, "neutral" = 2),
         Q2 = recode(Q2, "neutral" = 2, "agree" = 1),
         Q4 = recode(Q4, "Disagree" = 1),
         Q5 = recode(Q5, "Agree" = 3)
  )

You'll also want to read about the .default and .missing arguments in the help file to be sure you aren't introducing NAs when you don't mean to.

Answer 2

We can do this in base R without using any package. Create a lookup named vector ('v1'), loop over the columns and use that vector to change the values in the columns

v1 <- setNames(c(1:3, 3), c("Disagree", "neutral", "agree", "Agree"))
df1[-1] <- lapply(df1[-1], function(x) if(any(!is.na(x))) v1[x] else NA)
df1 
#  Participant Q1 Q2 Q3 Q4 Q5
#1           1  3  2 NA  1  3
#2           2  2  3 NA NA NA

data

df1 <- structure(list(Participant = 1:2, Q1 = c("agree", "neutral"), 
Q2 = c("neutral", "agree"), Q3 = c(NA, NA), Q4 = c("Disagree", 
NA), Q5 = c("Agree", NA)), .Names = c("Participant", "Q1", 
"Q2", "Q3", "Q4", "Q5"), class = "data.frame", row.names = c(NA, -2L))