Stepping Forward: Towards a More Systematic NPF with Automation

Abstract

Advancements in automated text analysis have substantially increased our capacity to study large volumes of documents systematically in policy process research. The Narrative Policy Framework (NPF)—which promotes empirical analysis of narratives—has the potential to usher policy narrative research along the same path. Using the NPF and existing semi-automated analysis tools, we investigate the relationship between narrative components—namely, characters and proposed solutions—and the more “skeletal” frames that tie policy narrative elements to one another. To illustrate how these tools can advance policy narrative research, we auto-code 5,708 state and local news articles focusing on hydraulic fracturing of oil and gas. The findings suggest that the use and role of characters and policy solutions are portrayed in significantly different ways depending on the frame used. By using an autocoding approach, these findings increase our methodological and theoretical understanding of the relationship between narrative elements and frames in policy narratives. In discussing these findings, we also consider their implications for how issue frames matter theoretically in the NPF.