Issue with reverse coding in R tidyr

Question

This is my code:*

data %>%
  mutate(r_stemster_3 = recode(stemster_3,
                               `1` = 7,
                               `2` = 6, 
                               `3` = 5,
                               `4` = 4, 
                               `5` = 3,
                               `6` = 2,
                               `7` = 1))

This is the error message:

Error in `mutate()`:
ℹ In argument: `r_stemster_3 = recode(...)`.
Caused by error in `recode()`:
! unused arguments (`1` = 7, `2` = 6, `3` = 5, `4` = 4, `5` = 3, `6` = 2, `7` = 1)
Backtrace:
 1. data %>% ...
 3. dplyr:::mutate.data.frame(...)
 4. dplyr:::mutate_cols(.data, dplyr_quosures(...), by)
 6. dplyr:::mutate_col(dots[[i]], data, mask, new_columns)
 7. mask$eval_all_mutate(quo)
 8. dplyr (local) eval()

Error in mutate(., r_stemster_3 = recode(stemster_3, `1` = 7, `2` = 6,  : 
  
Caused by error in `recode()`:
! unused arguments (`1` = 7, `2` = 6, `3` = 5, `4` = 4, `5` = 3, `6` = 2, `7` = 1)

I can't figure out what was wrong with my code. Please help fix the code!

I tried variations of the code such as removing the backticks or using single/double quotes instead of backticks. None of them worked.

Also data$stemster_3 is numeric and I want the reverse coded variable to be numeric too.

Hi, I believe that you may have a package that changed the definition of the recode function. Either load the tidyverse package before running the code or replace recode with dplyr::recode. If you are running both the car and dplyr package, this is most likely the issue — cravetheflame
– cravetheflame, Commented Feb 29, 2024 at 16:14
You are right @Cravetheflame. The issue lies in the packages. After I removed car, the codes worked perfectly. Thank you so much! — Shay
– Shay, Commented Feb 29, 2024 at 16:26
Happy to help, but please consider @Onyambu's answer as it demonstrates a simpler way of dealing with what you were trying to do :) — cravetheflame
– cravetheflame, Commented Feb 29, 2024 at 16:29
Note more "scalable" extension of Onyambu's answer would be (max(data$r_stemster_3)+1) - data$r_stemster_3 — jpsmith
– jpsmith, Commented Feb 29, 2024 at 16:42

MetehanGungor · Accepted Answer · 2024-02-29 19:09:05Z

There are many ways to deal with reverse coding in R. Here are some:

The simplest way

sampledf <- data.frame(V1 = c(1, 4, 5, 6, 7),
                       V2 = c(2, 4, 4, 5, 6),
                       V3 = c(2, 7, 1, 1, 5))
8 - sampledf

'psych' package

sampledf <- data.frame(V1 = c(1, 4, 5, 6, 7),
                       V2 = c(2, 4, 4, 5, 6),
                       V3 = c(2, 7, 1, 1, 5))

install.packages("psych")
library(psych)
reverseitems <- c(-1, 1, 1)
reverse.code(reverseitems, sampledf, mini = 1, maxi = 7)

note. reverseitems <- c(-1, 1, 1) is here to indicate V1 in the dataset. In this case, only the values in the first column will be reverse-coded.

'dplyr' package

sampledf <- data.frame(V1 = c(1, 4, 5, 6, 7),
                       V2 = c(2, 4, 4, 5, 6),
                       V3 = c(2, 7, 1, 1, 5))
library(dplyr)
sampledf |> 
  mutate(new = 8 - sampledf)

note. In this way, new variables will be added to the dataset (sampledf).

apply() function

sampledf <- data.frame(V1 = c(1, 4, 5, 6, 7),
                       V2 = c(2, 4, 4, 5, 6),
                       V3 = c(2, 7, 1, 1, 5))
sample123 <- apply(sampledf[, c(1, 2, 3)], 2, function(x) 8-x)
sample123

note. By using c(1, 2, 3) you can specify the variables (eg. all variables). After having new dataset, you can merge the specific variables of the old and new datasets (data.frame() function).

Collectives™ on Stack Overflow

Issue with reverse coding in R tidyr

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related