convert a column of categorical covariates into a number of columns with a binary flag for each category

Source: R/ordinal_to_binary.R

convert a column of categorical covariates into a number of columns with a binary flag for each category

ordinal_to_binary_(df, col_name, prefix = NULL, overwrite = FALSE)

Arguments

df

data frame
col_name

column name to split into multiple binary flags
prefix

string name if want to specify a prefix to label the binary flag columns other than the original col_name
overwrite

overwrite any existing columns if the newly generated columns share the same name

Examples

library(dplyr)
df <- data_frame(OCC = c(1, 1, 2, 3))
df %>% ordinal_to_binary_("OCC")
#>   OCC OCC1 OCC2 OCC3
#> 1   1    1    0    0
#> 2   1    1    0    0
#> 3   2    0    1    0
#> 4   3    0    0    1
df %>% ordinal_to_binary_("OCC", prefix = "OCCASION")
#>   OCC OCCASION1 OCCASION2 OCCASION3
#> 1   1         1         0         0
#> 2   1         1         0         0
#> 3   2         0         1         0
#> 4   3         0         0         1

df2 <- data_frame(OCC = c(1, 1, 2, 3), OCC1 = 999)
df2 %>% ordinal_to_binary_("OCC")
#> Warning: detected existing columns with same name sequence, 
#>             prepending "_" to all generated columns.
#>   OCC OCC1 _OCC1 _OCC2 _OCC3
#> 1   1  999     1     0     0
#> 2   1  999     1     0     0
#> 3   2  999     0     1     0
#> 4   3  999     0     0     1
df2 %>% ordinal_to_binary_("OCC", overwrite=T)
#>   OCC OCC1 OCC2 OCC3
#> 1   1    1    0    0
#> 2   1    1    0    0
#> 3   2    0    1    0
#> 4   3    0    0    1