Twitter

Existing text scaling methods often require a large corpus, struggle with short texts, or require labeled data. We develop a text scaling method that leverages the pattern recognition capabilities of generative large language models (LLMs). Specifically, we propose concept-guided chain-of-thought (CGCoT), which uses prompts designed to summarize ideas and identify target parties in texts to generate concept-specific breakdowns, in many ways similar to guidance for human coder content analysis. CGCoT effectively shifts pairwise text comparisons from a reasoning problem to a pattern recognition problem. We then pairwise compare concept-specific breakdowns using an LLM. We use the results of these pairwise comparisons to estimate a scale using the Bradley-Terry model. We use this approach to scale affective speech on Twitter. Our measures correlate more strongly with human judgments than alternative approaches like Wordfish. Besides a small set of pilot data to develop the CGCoT prompts, our measures require no additional labeled data and produce binary predictions comparable to a RoBERTa-Large model fine-tuned on thousands of human-labeled tweets. We demonstrate how combining substantive knowledge with LLMs can create state-of-the-art measures of abstract concepts.

There is widespread concern that foreign actors are using social media to interfere in elections worldwide. Yet data have been unavailable to investigate links between exposure to foreign influence campaigns and political behavior. Using longitudinal survey data from US respondents linked to their Twitter feeds, we quantify the relationship between exposure to the Russian foreign influence campaign and attitudes and voting behavior in the 2016 US election. We demonstrate, first, that exposure to Russian disinformation accounts was heavily concentrated: only 1% of users accounted for 70% of exposures. Second, exposure was concentrated among users who strongly identified as Republicans. Third, exposure to the Russian influence campaign was eclipsed by content from domestic news media and politicians. Finally, we find no evidence of a meaningful relationship between exposure to the Russian foreign influence campaign and changes in attitudes, polarization, or voting behavior. The results have implications for understanding the limits of election interference campaigns on social media.