CODEWITHSUNDEEP
rnumeric
Pre
Pre

Reputation: 111

Combining two numeric variables with unite and retaining new column as a numeric variable in r

In trying to use unite to concatenate two numeric columns, I think this is turning the resulting column into a character string - how to avoid this?

df  %>% unite(housinggrossconcat, housinggrossedupnet, housing.gross, na.rm=TRUE, remove=FALSE))
                %>% unite(grossconcat, grossedupnet, gross, na.rm=TRUE, remove=FALSE)

Data looks like this:

Gross    Grossedupnet    New Column: Grossconcat
 30         NA              30
 NA         45              45 
 NA         45              45
 350        NA              350

It is such that wherever Gross has a value, Grossedupnet will be NA, and vice versa. They're both numerical values. I want to concatenate the two into the new column, but it is turning the new column into a character variable.

Upvotes: 0

Views: 1355

Answers (2)

akrun
akrun

Reputation: 887158

We can also use fcoalesce

library(data.table)
setDT(df)[, Grossconcat := fcoalesce(Gross, Grossedupnet)]

Upvotes: 0

Gregor Thomas
Gregor Thomas

Reputation: 145785

You're looking for coalesce:

df %>%
  mutate(Grossconcat = coalesce(Gross, Grossedupnet))

dplyr::coalesce() is built for the purpose of taking the first non-missing value. It works on as many columns as you give it, and it works well for all data types. unite is the complement to separate. It is for pasting together strings, hence the character output.

Upvotes: 1

Related Questions