When importing a dataset from Stata to R, it often comes with helpful labels for numeric variables. I would like to be able to convert the data in the labels to a new separate variable. The equivalent command in Stata is decode.
library(tidyverse)
library(webuse)
auto <- webuse("auto")
auto$foreign #Want to convert this to a character variable that reads "Domestic" or "Foreign"
One option is to use the labelled package, e.g.
library(tidyverse)
#install.packages("webuse")
library(webuse)
#install.packages("labelled")
library(labelled)
auto <- webuse("auto")
auto$foreign
auto$labels <- labelled::to_factor(auto$foreign, levels = "labels")
auto$labels
#>[1] Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic
#>[13] Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic
#>[25] Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic
#>[37] Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic
#>[49] Domestic Domestic Domestic Domestic Foreign Foreign Foreign Foreign Foreign Foreign Foreign Foreign
#>[61] Foreign Foreign Foreign Foreign Foreign Foreign Foreign Foreign Foreign Foreign Foreign Foreign
#>[73] Foreign Foreign
#>attr(,"label")
#>[1] Car type
#>Levels: Domestic Foreign
Or, to keep the values as well as the labels:
auto$labels <- labelled::to_factor(auto$foreign, levels = "prefixed")
auto$labels
#>[1] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic
#>[9] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic
#>[17] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic
#>[25] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic
#>[33] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic
#>[41] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic [0] Domestic
#>[49] [0] Domestic [0] Domestic [0] Domestic [0] Domestic [1] Foreign [1] Foreign [1] Foreign [1] Foreign
#>[57] [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign
#>[65] [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign [1] Foreign
#>[73] [1] Foreign [1] Foreign
#>attr(,"label")
#>[1] Car type
#>Levels: [0] Domestic [1] Foreign
To use dplyr mutate:
library(tidyverse)
#install.packages("webuse")
library(webuse)
#install.packages("labelled")
library(labelled)
auto <- webuse("auto")
auto %>%
mutate(labels = labelled::to_factor(auto$foreign, levels = "labels")) %>%
select(labels)