R create new variable if other variable contains vector-CodePudding

I want a create a treatment variable that takes the value of 1 for treated countries and 0 otherwise. I know how to do this individually for each country with transform but I was wondering if I can also add a vector to make it faster? I tired the following but I got an error (Warning: longer object length is not a multiple of shorter object length)

countries_vector <- c("USA", "UK", "ESP", "FR", "ITA")

df <- transform(df, treated = ifelse(Country == countries_vector, 1, 0))

CodePudding user response：

You can use mutate, ifelse, and %in% to quickly assign 0 or 1.

library(dplyr)
treated_countries <- c("USA", "UK", "ESP", "FR", "ITA")
all_countries <- sample(x = c("USA", "UK", "ESP", "FR", "ITA", "BRA", "JAP", "CHN", "ZAF", "DEU"),
                  size = 100, replace = TRUE)
df <- as.data.frame(all_countries)
df <- df %>%
  mutate(treated = ifelse(all_countries %in% treated_countries, 1, 0))

df %>%
  group_by(treated, all_countries) %>%
  summarize()
#> `summarise()` has grouped output by 'treated'. You can override using the
#> `.groups` argument.
#> # A tibble: 10 × 2
#> # Groups:   treated [2]
#>    treated all_countries
#>      <dbl> <chr>        
#>  1       0 BRA          
#>  2       0 CHN          
#>  3       0 DEU          
#>  4       0 JAP          
#>  5       0 ZAF          
#>  6       1 ESP          
#>  7       1 FR           
#>  8       1 ITA          
#>  9       1 UK           
#> 10       1 USA