KEY PUBLICATIONS The proliferation of algorithms is reconfiguring our socio- economic systems in profound and complex ways. This collaborative project, led by Martin Wells (Statistics and Data Science), brings together scholars from across the social sciences to critically examine the design, understanding, and use of algorithmic systems. The researchers share a concern with how the concomitant rise of big data, machine learning, and digital surveillance has the potential to exacerbate social inequalities among vulnerable communities. The proposed program of research will therefore develop a more holistic understanding of algorithmic bias across disciplinary boundaries and within such empirical domains as work/ employment, finances/creditworthiness, health, and social media. Through such research, we aim to understand— and ultimately challenge—the kinds of data-driven inequality and discrimination that are defining social life in the algorithmic age. Ifeoma Ajunwa (ILR School) and Martin Wells’ Mapping the Eco-system of Hiring Platforms project will interview the developers of hiring platforms and create a network view of the end users and saturation by industry/field. Wells and colleagues are developing statistical methods for estimation of causal effects from observational data and are applying these methods to assessing various fairness measures in machine learning. This approach to causal inference formulates the distinction between associational and causal concepts in terms of underlying joint distributions. Solon Barocas (Information Science) along with Cornell graduate student Manish Raghavan (Computer Science) and Professors Karen Levy (Information Science) and Jon Kleinberg (Computer and Information Science) are investigating the practices of an emerging set of companies that specialize in algorithmic pre-employment assessments, with the goal of establishing what companies mean when they assert that their tools reduce or eliminate bias from the hiring process. Brooke Erin Duffy (Communication) and Cornell students Annika Pinch (Psychology) and Shruti Sannon (Communication) are collaborating on a study examining the impact of algorithmic systems on work in the media and creative industries. Through in-depth interviews with online content creators and gig workers, their study shows how cultural producers respond to the demands of algorithmic learning to achieve efficiencies of work. Malte Ziewitz (Science & Technology Studies) and organization studies scholar Maximilian Heimstädt have been following recent attempts in New York City to introduce legislation for algorithmic accountability. In 2018, Ajunwa and Wells received an ILR School theme grant on Technology and Work ($20K). Ajunwa’s NSF CAREER Award proposal, “The Development, Design, and Ethical Issues of Algorithmic Hiring Tools” was recommended for funding in the amount of $526,877. Ziewitz received an NSF CAREER Award for a project on “Understanding and Advancing Fair Representation in Algorithmic Systems” ($400,300). Ziewitz and Stephen Hilgartner have been awarded $178,361 by the Cornell Data Science Curriculum Initiative to establish the “Data Science & Society Lab.” Wells’ research has been supported by an NSF grant ($150,000) “Variable Selection When P>>N: Beyond the Linear Regression and Normal Errors Model” and a Department of Labor grant ($244,603) “Criminal Record Inaccuracies and the Impact of a Record Education Intervention on Employment-Related Outcomes.” Barocas and Cornell colleagues received a grant from the MacArthur Foundation to support a new initiative on Artificial Intelligence, Policy, and Practice ($900,000). He and colleagues were also awarded a NSF grant for a new project on “Emerging Cultures of Data Science Ethics in the Academy and Industry” ($401,914). NEW FUNDING Ifeoma Ajunwa Age Discrimination by Algorithms Berkeley Journal of Employment and Labor Law Algorithms at Work: Productivity Monitoring Applications and Wearable Technology St. Louis Law Journal Samir Passi and Solon Barocas Problem Formulation and Fairness Conference on Fairness, Accountability, and Transparency, 2019 Andrew Selbst and Solon Barocas The Intuitive Appeal of Explainable Machines Fordham Law Review Brooke Erin Duffy, et al. Imagining and Resisting Algorithmic Change: Networked Creative Communities Presented at the 69th Annual Conference of the International Communication Association Martin Wells, et al. FacilitatingHigh-DimensionalTransparentClassification Via Empirical Bayes Variable Selection Applied Stochastic Models in Business and Industry Malte Ziewitz Rethinking Gaming: The Ethical Work of Optimization in Web Search Social Studies of Science KEY PUBLICATIONS 7 2018-2021 ALGORITHMS, BIG DATA, AND INEQUALITY