Individual Submission Summary
Share...

Direct link:

Group Appeals and the Origins of Party Reputations

Thu, September 5, 12:00 to 1:30pm, Loews Philadelphia Hotel, Commonwealth C

Abstract

Party reputations, widely shared perceptions of what political parties stand for and who they represent, are an essential feature of party competition. Earlier scholarship identifies parties' connections to social groups as an important constituent cause of party reputations and thereby voter behavior. How do parties establish and maintain such connections? We study the role of group appeals, i.e. valenced references to social groups in public speech. We theorize that parties strategically use group appeals to forge and maintain linkages to social groups, which in turn underpins party reputations. This mechanism is often assumed in the group appeals literature but has rarely been tested, and never outside of experimental settings.

To test this theory, we take a novel approach to measuring group appeals in large bodies of text. A common limitation of existing analyses is the assumption that group appeals are equivalent to group mentions. However, many group mentions are not valenced, and some are even negative. Moreover, the valence of an appeal is rarely reflected in sentence sentiment.

To overcome these problems and accurately measure group appeals, we train a large language model to evaluate each group mention as if it were a member of the group. After extensive validation of the model's performance against human coding, we apply this measure to group mentions across millions of sentences to capture trends in group appeals within dozens of group-party dyads in two parliamentary systems.

We link the resulting data to panel surveys measuring citizens' changing perceptions of each group-party linkage over several decades. With these linked data, we examine whether and how party elites use group appeals to shape party reputations over time. In a set of follow-up analyses, we hold constant the role of policy, isolating the effect of the purely symbolic dimension of group appeals.

Our findings contribute to the existing literature on multiple fronts. First, we advance the measurement of group appeals, a key concept in party politics. Second, we shed new light on the role of group appeals in shaping public opinion and how they underpin party reputations. Third, and more generally, we move beyond the conventional static understanding of party reputations to study temporal dynamics of party reputations.

Authors