Public Values and Technological Change: Mapping how Municipalities Grapple with Data Ethics

Local governments in the Netherlands are increasingly undertaking data projects for public management. While the emergence of data practices and the application of algorithms for decision making in public management have led to a growing critical commentary, little actual empirical research has been conducted. Over the past few years, we have developed a research method that enables researchers to enter organisations not merely as researchers but also as experts on data ethics. Through participatory and ethnographic observation, the DEDA (Data Ethics Decision Aid) gives us special insight into ethics in local government. Where most research has focused on the theoretical aspects of data ethics, our approach offers a new perspective on data practices, by looking at how data ethics is done in public management. Our research provides insight into the state of data awareness within organisations that are mostly portrayed—within critical data studies—as homogeneous and monolithic entities. The distinct method developed at Utrecht Data School allows researchers to immerse themselves within organisations and closely observe data practices, discourses on ethics, and how organisations address challenges that arise as a consequence of datafication. For the purpose of this chapter, we analyse our field work with the DEDA through the lens of Mark Moore’s Strategic Triangle of Public Value. We show how our field work can give insight into how the three angles of the strategic triangle are shaped in practice. From this analysis we draw three conclusions. First, that ethical awareness of data projects is often low because data literacy among civil servants is limited. Second, that by not recognising the choices civil servants have to make as ethical or political choices, they can make decisions that go beyond their mandate. Third, that there is a dangerous tendency where ethical deliberation is sometimes seen as an obnoxious bureaucratic box ticking exercise, instead of being considered as a vital part of the design and build-up of a data project.