Chatbot-Driven Voting Aid Applications Increase Knowledge about Party Positions but Do Not Change Party Evaluations

Voting Aid Applications (VAAs) are interactive tools that communicate information about elections, yet their effectiveness in enhancing political knowledge and participation remains understudied. Moreover, traditional VAAs may disproportionately attract politically engaged users with already well-formed ideological views, limiting their potential to inform a broader and less engaged electorate. This paper introduces a novel “VAA Bot” that employs large language models and retrieval-augmented generation to deliver balanced, personalized information drawn from official party documents. We evaluate the VAA Bot’s impact across three experimental studies aimed at young adults. The findings provide evidence that the VAA Bot improves knowledge of party stances. However, we observe weaker effects on downstream outcomes such as vote preferences and party evaluations. These findings contribute to ongoing debates about the role of political information in shaping behavior and underscore both the promise and the limitations of LLM-based tools for civic learning.