Getting a NA for sum across rows if any variable values is NA-CodePudding

I have a dataset with var1, var2, var3, var4, and I am calculating a sum var_total \<- var1 var2 var3 var4. I want missing value in var_total if any of the values var1, var2, var3 and var4 is missing.

Have:

Var1	var2	var3	var4	var_total
1	0	0	0	1
1	NA	2	0	3
1	0	0	NA	1

Want:

Var1	var2	var3	var4	var_total
1	0	0	0	1
1	NA	2	0	NA
1	0	0	NA	NA

I assume something involving ifelse().

CodePudding user response：

Libraries

library(dplyr)

Data

data <-
  tibble::tribble(
    ~var1, ~var2, ~var3, ~var4, ~var_total,
    1L,    0L,    0L,    0L,         1L,
    1L,    NA,    2L,    0L,         3L,
    1L,    0L,    0L,    NA,         1L
  )

Code

data %>% 
  rowwise() %>% 
  mutate(var_total = sum(c_across(cols = var1:var4),na.rm = FALSE))

Output

# A tibble: 3 x 5
# Rowwise: 
   var1  var2  var3  var4 var_total
  <int> <int> <int> <int>     <int>
1     1     0     0     0         1
2     1    NA     2     0        NA
3     1     0     0    NA        NA

CodePudding user response：

Using rowSums() in base R:

data$var_total <- rowSums(data[ , 1:4])

Or with dplyr:

library(dplyr)

data %>%
  mutate(var_total = rowSums(across(var1:var4)))

Result from either approach:

  var1 var2 var3 var4 var_total
1    1    0    0    0         1
2    1   NA    2    0        NA
3    1    0    0   NA        NA