Algorithmic Decision Making in Public Administration

Abstract

In this paper, I propose a study of algorithmic decision making in public administration from a computer supported cooperative work (CSCW) perspective. Each day the public administration makes thousands of decisions with consequences for the welfare of its citizens. An increasing number of such decisions are supported or made by algorithmic decision making (ADM) systems, yet in the scientific and public sphere there is a growing concern that these algorithms become a 'black box' possibly containing hidden bias (Olsen et al., 2019), obstacles for human discretion (Rason, 2017), low transparency (Alkhatib and Bernstein, 2019) or trust (Mittelstadt et al. 2016). For example, ADM is currently tested in public administration in job placement for the prediction of a citizen's risk of long-term unemployment. Following prior research questioning the usefulness of the black box metaphor, my interest is to understand how caseworkers' and citizens understand ADM, as a basis for design of CSCW technologies employing ADM.