Skip to content

⚤ Get sex (female percent) estimates based on first names

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md
Notifications You must be signed in to change notification settings

mkearney/name2sex

Repository files navigation

⚤ name2sex

lifecycle

The goal of name2sex is to convert a vector of first names into sex estimates (female percent) based on historical data bases.

NOTE

If you’re reading this, I’d encourage you to check out https://github.com/ropensci/gender instead. In making this package, we made some decisions that are probably not completely obvious. So it’d be more transparent [and probably all-around better] to use the https://github.com/ropensci/genderdata and https://github.com/ropensci/gender packages directly!

Installation

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("mkearney/name2sex")

Example

This is a basic example:

## load package
library(name2sex)

## vector of first names
first_names <- c("casi", "mike", "tracy", "rory", "pat", "alex", "tamar")

## apply names2sex function to first names
names2sex(first_names)
#>       casi       mike      tracy       rory        pat       alex 
#> 1.00000000 0.00313833 0.80361373 0.16616643 0.58198674 0.02903625 
#>      tamar 
#> 0.86616352

About

⚤ Get sex (female percent) estimates based on first names

Topics

Resources

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages