case_when inserting NAs instead of new assigned values despite all levels are included

[GabryS3]

Hi,
I have been using case_when without any issue so far.
However, I am having an issue at the moment and can;t figure out if it's a bug or something i am doing wrong, as I receive no error from running the function.

So, I have a dataframe with the following column "Location" with 6 levels:

class(Metadata) # data.frame
class(Metadata$"Catch Location") # factor

levels(Metadata$"Catch Location") 
[1] "1A (Moreton North)"  "A2 (Moreton South)"  "B1 (Cooktown North)" "B2 (Cooktown South)"    "C1 (GOC North)"    "C2 (GOC South)" 

When I try to add a new field (column) called "pop" with mutate and case_when, I get all NAs in the pop column, why?
I use the following code:

Metadata = Metadata %>% mutate(pop = (case_when("Catch Location" == "1A (Moreton North)" ~ "1A_Moreton North",
                                                           "Catch Location" == "A2 (Moreton South)" ~ "A2_Moreton South",
                                                           "Catch Location" == "B1 (Cooktown North)" ~ "B1_Cooktown North",
                                                           "Catch Location" == "B2 (Cooktown South)" ~ "B2_Cooktown South",
                                                           "Catch Location" == "C1 (GOC North)" ~ "C1_GOC North",
                                                           "Catch Location" == "C2 (GOC South)" ~ "C2_GOC South")))

Why is it giving me all NAs in the pop field? I have used the exact same code before withou isssues, so I am not sure what is happening.
I tried with Location as "character" or "factor" and it makes no difference...always same issue...NAs in all pop column.

I would appreciate any feedback and help!
Thanks!
Gabry

[joranE]

I think you need to put Catch Location on the LHS in backticks not quote marks. i.e.

case_when(`Catch Location` == 

Otherwise it's interpreting the LHS as the literal string "Catch Location" which certainly is never equal to any of the values you compare it to.

...or avoid using column names with spaces in them so you can refer to the columns as bare column names.

[DavisVaughan]

Yea @joranE is right here!

[GabryS3]

Thank you so much! that fixed it!!