Predict the race of a surname

Jacob Kaplan

2020-07-21

library(predictrace)

Usage

The only required parameter is name which is the surname you want to find the race of. Please note that this parameter only accepts surnames, including both first and last name will result in not finding a match in the Census data.

predict_race("Washington")
#>         name match_name likely_race probability_american_indian
#> 1 Washington washington       black                      0.0068
#>   probability_asian probability_black probability_hispanic probability_white
#> 1             0.003            0.8753               0.0254            0.0517
#>   probability_2races
#> 1             0.0378

This function accepts a single string or a vector of strings.

predict_race(c("Washington", "Franklin", "Lincoln"))
#>         name match_name likely_race probability_american_indian
#> 1 Washington washington       black                      0.0068
#> 2   Franklin   franklin       white                      0.0083
#> 3    Lincoln    lincoln       white                      0.0368
#>   probability_asian probability_black probability_hispanic probability_white
#> 1            0.0030            0.8753               0.0254            0.0517
#> 2            0.0054            0.3876               0.0270            0.5438
#> 3            0.0135            0.1471               0.0251            0.7490
#>   probability_2races
#> 1             0.0378
#> 2             0.0278
#> 3             0.0285

If you only want the most likely race and not the individual probabilities of each race, set the parameter probability to FALSE.

predict_race("Washington", probability = FALSE)
#>         name match_name likely_race
#> 1 Washington washington       black