How to convert a variable with yes or no values to 1 and 0 in R?-CodePudding

I'm working on an employee attrition dataset. There is a column named attrition and it contains 1470 yes and no values. I want to convert these yes and no values to 1 and 0. I installed the dplyr package in R to use the mutate function but I'm having trouble converting. Here's what I tried:

clean %>% mutate(empAttrcopy$attrition=ifelse(empAttrcopy$attrition=="NO",0,1)) Error: unexpected '=' in "clean %>% mutate(empAttrcopy$attrition="

CodePudding user response：

Assuming attrition contains only "YES" or "NO" values, this should work. First some reproducible data:

set.seed(42)
attrition <- sample(c("YES", "NO"), 15, replace=TRUE)
attrition
#  [1] "YES" "YES" "YES" "YES" "NO"  "NO"  "NO"  "NO"  "YES" "NO"  "YES" "NO"  "YES" "NO"  "YES"
attrition <- as.numeric(attrition=="YES")
attrition
#  [1] 1 1 1 1 0 0 0 0 1 0 1 0 1 0 1

I've left off the empAttrcopy$ prefix for simplicity.

CodePudding user response：

I suspect you have a dataframe called "empAttrcopy", want to dummy code the column "attrition" and store it to a new dataframe called "clean".

When using dplyr, the pipe %>% (originally from magrittr) evaluates what's on the left side of it and feeds the result as input to the next function on the right. So if you want to mutate a column within "empAttrcopy", you need to put it before the first %>%. When having done that, R knows that you are working with "Attrition", so there's no need to backreference it after piping (i.e. no need for $).

empAttrcopy %>% mutate(attrition = ifelse(attrition == "No", 0, 1))

If you want to store the output of your mutated column to a new dataframe called "clean", do so by assigning it by using <- at the start or -> at the end.

clean <- empAttrcopy %>% mutate(attrition = ifelse(attrition == "No", 0, 1))

# or

empAttrcopy %>% mutate(attrition = ifelse(attrition == "No", 0, 1)) -> clean

Last but not least, it's always better to store the output in a new column (e.g. save it to "attrition_dummy"), so that you don't lose any data.

clean <- empAttrcopy %>% mutate(attrition_dummy = ifelse(attrition == "No", 0, 1))

CodePudding user response：

You can just use the ifelse function available on the base package. This allows an expression to be evaluated and assigned values as appropriate.

a = c(5,7,2,9)
(a %% 2 == 0,"even","odd")

[1] "odd" "odd" "even" "odd"