Search

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., Comment: 26 pages, 2 figures

Existing approaches to estimating politicians' latent positions along specific dimensions often fail when relevant data is limited. We leverage the embedded knowledge in generative large language models (LLMs) to address this challenge and measure lawmakers' positions along specific political or policy dimensions. We prompt an instruction/dialogue-tuned LLM to pairwise compare lawmakers and then scale the resulting graph using the Bradley-Terry model. We estimate novel measures of U.S. senators' positions on liberal-conservative ideology, gun control, and abortion. Our liberal-conservative scale, used to validate LLM-driven scaling, strongly correlates with existing measures and offsets interpretive gaps, suggesting LLMs synthesize relevant data from internet and digitized media rather than memorizing existing measures. Our gun control and abortion measures -- the first of their kind -- differ from the liberal-conservative scale in face-valid ways and predict interest group ratings and legislator votes better than ideology alone. Our findings suggest LLMs hold promise for solving complex social science measurement problems., Comment: 18 pages, 4 figures; V2: fixed graphical error on Figure 2; V3: reorganized sections, updated prose; V4: added additional scales and analysis

Text analysis in the social sciences often involves using specialized dictionaries to reason with abstract concepts, such as perceptions about the economy or abuse on social media. These dictionaries allow researchers to impart domain knowledge and note subtle usages of words relating to a concept(s) of interest. We introduce the dictionary-assisted supervised contrastive learning (DASCL) objective, allowing researchers to leverage specialized dictionaries when fine-tuning pretrained language models. The text is first keyword simplified: a common, fixed token replaces any word in the corpus that appears in the dictionary(ies) relevant to the concept of interest. During fine-tuning, a supervised contrastive objective draws closer the embeddings of the original and keyword-simplified texts of the same class while pushing further apart the embeddings of different classes. The keyword-simplified texts of the same class are more textually similar than their original text counterparts, which additionally draws the embeddings of the same class closer together. Combining DASCL and cross-entropy improves classification performance metrics in few-shot learning settings and social science applications compared to using cross-entropy alone and alternative contrastive and data augmentation methods., Comment: 6 pages, 5 figures, EMNLP 2022

To mitigate the spread of fake news, researchers need to understand who visit fake new sites, what brings people to those sites, where visitors come from, and what content they prefer to consume. In this paper, we analyze web traffic data from The Gateway Pundit (TGP), a popular far-right website that is known for repeatedly sharing false information that has made its web traffic available to the general public. We collect data on 68 million web traffic visits to the site over a month period and analyze how people consume news via multiple features. Our traffic analysis shows that search engines and social media platforms are main drivers of traffic; our geo-location analysis reveals that TGP is more popular in counties that voted for Trump in 2020; and our topic analysis shows that conspiratorial articles receive more visits than factual articles. Due to the inability to observe direct website traffic, existing research uses alternative data source such as engagement signals from social media posts. To validate if social media engagement signals correlate with actual web visit counts, we collect all Facebook and Twitter posts with URLs from TGP during the same time period. We show that all engagement signals positively correlate with web visit counts, but with varying correlation strengths. Metrics based on Facebook posts correlate better than metrics based on Twitter. Our unique web traffic data set and insights can help researchers to better measure the impact of far-right and fake news URLs on social media platforms.

In January 2019, YouTube announced it would exclude potentially harmful content from video recommendations but allow such videos to remain on the platform. While this step intends to reduce YouTube's role in propagating such content, continued availability of these videos in other online spaces makes it unclear whether this compromise actually reduces their spread. To assess this impact, we apply interrupted time series models to measure whether different types of YouTube sharing in Twitter and Reddit changed significantly in the eight months around YouTube's announcement. We evaluate video sharing across three curated sets of potentially harmful, anti-social content: a set of conspiracy videos that have been shown to experience reduced recommendations in YouTube, a larger set of videos posted by conspiracy-oriented channels, and a set of videos posted by alternative influence network (AIN) channels. As a control, we also evaluate effects on video sharing in a dataset of videos from mainstream news channels. Results show conspiracy-labeled and AIN videos that have evidence of YouTube's de-recommendation experience a significant decreasing trend in sharing on both Twitter and Reddit. For videos from conspiracy-oriented channels, however, we see no significant effect in Twitter but find a significant increase in the level of conspiracy-channel sharing in Reddit. For mainstream news sharing, we actually see an increase in trend on both platforms, suggesting YouTube's suppressing particular content types has a targeted effect. This work finds evidence that reducing exposure to anti-social videos within YouTube, without deletion, has potential pro-social, cross-platform effects. At the same time, increases in the level of conspiracy-channel sharing raise concerns about content producers' responses to these changes, and platform transparency is needed to evaluate these effects further.

Despite many examples to the contrary, most models of elections assume that rules determining the winner will be followed. We present a model where elections are solely a public signal of the incumbent popularity, and citizens can protests against leaders that do not step down from power. In this minimal setup, rule-based alternation of power as well as "semi-democratic" alternation of power independent of electoral rules can both arise in equilibrium. Compliance with electoral rules requires there to be multiple equilibria in the protest game, where the electoral rule serves as a focal point spurring protest against losers that do not step down voluntarily. Such multiplicity is possible when elections are informative and citizens not too polarized. Extensions to the model are consistent with the facts that protests often center around accusations of electoral fraud and that in the democratic case turnover is peaceful while semi-democratic turnover often requires citizens to actually take to the streets., Comment: 41 pages, 3 figures