This project aims to make it easier for HR/people analysts to compare the representation of different demographics in their company to the communities they have a presence in. It also helps identify if a company is getting the application numbers from different groups they would expect based on the same criteria.

Especially with very distributed companies, it is often difficult to see which areas you have the biggest struggles in attracting and retaining talent from different backgrounds. This project helps make that process a bit easier, by creating a framework for these comparisons to be made.

Download the project and unzip it. Open the R file of interest (USA, International, Technical USA), and import the data sources from your company as instructed. This project will calculate the representation numbers of a particular demographic (starting with gender) across different geographic regions (starting with states in the USA or Countries across the world). Then, it will take the labour market data I’ve compiled to estimate the representation of that demographic group in the labour force in that region. Finally, it calculates the difference between the expected and actual representations of underrepresented groups, and writes a CSV file that highlights areas where your company has a significant presence and has a lot less of a particular group than one would expect. If you want to add the optional application rates piece, it goes through a similar process to help you see if you have issues attracting talent from underrepresented groups at the top of the funnel.

Labor force participation and demographics by state comes from the American Community Survey for Gender Across USA (note: labor force participation rates are from 2013, demographics from 2017). For Gender Across World, female labor force percentage comes from International Labour Organization, ILOSTAT database (2017). For Technical Gender Across USA, they come from a specific aggregation of the American Community Survey and are from 2013.

This is the case because these are the only area where I personally was able to find the data I wanted. This was always meant to be a starting place, and I’m looking forward to the open source process in helping this project advance.

Check out the “Contributing” file; your help is much appreciated!

I do work at GitHub, and it’s awesome! While GitHub is supportive of me contributing to open source, this project is independent of my work there, and in no way reflects GitHub’s philosophy or projects on this topic. This is just something I find personally interesting and want to build with the community.