Policy Analytics: Definitions, Components, Methods, and Illustrative Examples

Abstract

There are many ways to define policy analytics. In fact, different terms are used for what could be considered the same phenomenon: the use of data and analytical techniques to make policy decisions. Policy analytics, policy modelling, and policy informatics are just some of the most used terms by scholars and practitioners. There is no clarity, however, in terms of where the boundaries of this concept lie and what main analytical methods it includes. First, regarding the conceptual boundaries, some experts argue that policy analytics includes analysis-related tasks only and has nothing to do with data preparation, management, governance, and stewardship. For other experts, policy analytics encompasses all the activities in the data lifecycle and includes elements outside of the analytic sphere, such as information technologies, stakeholder involvement, and a deep understanding of the context of use and application domains. Second, in terms of analytical methods, some analysts consider data mining, machine learning, and other computer science approaches to be the only valid tools. For others, statistical analysis as well as simulation approaches should also be called analytics. In this chapter, we propose a comprehensive and integrative view in which policy analytics goes beyond data analysis and includes management and preparation of data as well as very diverse techniques such as computer simulation, social network analysis, statistics, geographic information systems, and data mining techniques.