The study of policy coordination: an approach to integrate expert assessment with automated content analysis

Abstract Public policy scholars have long agreed that the coordination of policies involving different authorities and policy frameworks is challenging. This paper responds to recent calls to explore new empirical methods for studying policy coordination by integrating automated text analysis into the workflow of human experts. A key contribution of our approach is a quantitative indicator of the semantic alignment between different policy objectives. We critically review the performance of the indicator in terms of its ability to assess both horizontal and vertical policy coordination by analysing the Recovery and Resilience Plan (2021–26) and the national and regional Smart Specialization Strategies (2021–27) in Portugal. Based on this exercise, we reflect on the necessary conditions for the successful implementation of automated text analysis of policy documents as well as its potential in other empirical settings.