New R Package: Data Science Looks at Discrimination (dsld)

I’m very pleased to announce a new package, dsld, available on CRAN. This is the work of eight talented undergrad students. I provided the concept and some general guidance, but this is their work.

The package is aimed at dealing with discrimination — race, gender, age — in the workplace, education, health care and so on. It consists of analytical, graphical and tabular tools for:

• detection of discrimination
• addressing discriminatory predictive models

We hope the package will be useful in a variety of application venues, such as:

• inspiring motivation in statistics courses
• discrimination litigation
• internal HR audits
• social science research

The usefulness of the package is further enhanced by the availability of a free companion textbook. It uses dsld examples throughout, but its role is to explain the statistical concepts, not to serve as a user manual for the package.

One can acquire a good idea of the nature of the book by reading the example on law school admissions, pp. 34-43.

Needless to say, comments are welcome!