Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 66 - Meeting Client Needs: Approaches and Examples
Type: Contributed
Date/Time: Sunday, August 7, 2022 : 4:00 PM to 5:50 PM
Sponsor: Section on Statistical Consulting
Abstract #320926
Title: Equitable Random Assignment
Author(s): Clark Fitzgerald*
Companies: Sacramento State University
Keywords: equity; diversity; algorithm; software; randomization; workload
Abstract:

Suppose you are a member of a committee charged with reviewing many applications. How do you randomly assign committee members to applications as equitably as possible?

In this talk, we define what equitable random assignment means, explain why it’s desirable, and propose an algorithm to solve the problem. More generally, the problem is to randomly assign a combination of workers to tasks such that each worker has the same load, and each task is equally likely to have any combination of workers. We have incorporated equitable random assignment in our department’s hiring process to improve equity, which was the original motivation for this work. We conclude with a demonstration of our freely available open source software implementation, which is ready for public use.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program