Representativeness in Statistics, Politics, and Machine Learning

Explore the concept of representativeness across statistics, politics, and machine learning in this 19-minute conference talk from the FAccT 2021 virtual event. Delve into the statistical roots of representation, its application in science and fact, and the normative tensions that arise. Examine case studies such as crop reporting and exclusionary representativeness to gain a comprehensive understanding of this crucial topic in computing and society.

Introduction
Statistical Roots
Representation in Science
Representation in Fact
Normative Tensions
Crop Reporting
Exclusionary Representativeness
Summary