Using LLMs to Detect Group Appeals in Parties’ Election Manifestos

A growing political science scholarship substantively distinguishes between parties’ group appeals and policy appeals, the latter referring to the policies parties claim to support or oppose and the former relating to the specific social groups mentioned in party statements. While both are prevalent in parties’ campaign materials, existing large-scale datasets focus primarily on policy-based appeals, with little data available for empirical analyses of group-based appeals. Moreover, the extensive resources needed to generate these data via hand-coding and the complexity of automating such text analysis, provides challenges to new collection efforts. However, examining group appeals is substantively important for understanding party behavior, representation, and electoral politics. In this paper, we propose to address this gap by automating the extraction of group appeals from political texts using recent advances in computational text analysis to identify the social groups mentioned and the stance towards these groups (whether the appeals are supportive or critical). We use manifestos of major British, Irish, German and Austrian parties as test cases, demonstrating the benefits of our approach for revealing important trends in party behavior, with implications for party competition, voter behavior, and representation.