#R how to mutate case_when based on column name taken from vector

Question

I'm working to create factor variables from numeric column, but i want to use listed value as parameter for desired column. I have been able to select a column based on vector input but still unable to do it inside mutate case_when

var1 = "transfer"
var2 = "transfer_tier"

tier <- trx %>%
  select_(.dots = var1) %>% 
  mutate_(var2 = case_when(
    var1 <= 1e3 ~ '1. <1k',
    var1 <= 1e4 ~ '1. <10k',))

using above code shows no error but the formula that i want does not work. var2 is created with all NA. how to call var1 properly?

score 1 · Accepted Answer · answered Dec 01 '21 at 01:41

I can't reproduce your issue because I don't have your data, so here is a minimal reproducible example showing how to use as.name() and/or the "curly curly" syntax to bring names from 'external' vectors in your tidyverse workflow:

library(tidyverse)
#install.packages("palmerpenguins")
library(palmerpenguins)

var1 <- "body_mass_g"
var2 <- "island"

penguins %>%
  select(external_var1 = as.name(var1),
         external_var2 = {{var2}}) %>%
  mutate(categorised_var1 = case_when(
    external_var1 <= 3500 ~ "skinny",
    external_var1 >= 5500 ~ "heavy",
    TRUE ~ "regular"
  )) %>%
  ggplot(aes(y = categorised_var1, fill = external_var2)) +
  geom_bar()

^{Created on 2021-12-01 by the reprex package (v2.0.1)}

In this case as.name() and curly curly syntax are interchangeable, but that might not be the case with your real data; please see How to make a great R reproducible example for tips on how to help us help you.

That's the answer that I'm looking for, really appreciate it. also with as.name is more neater and elegant for me compared to.dots for selection from vector — Anshari, Dec 01 '21 at 02:05

#R how to mutate case_when based on column name taken from vector

1 Answers1