DISRUPTive effects of social media: politics, collective action and the economy – DISRUPT

The collection and analysis of online campaign data are central to understanding the influence of social media. A (semi-)automated process has been developed to collect content from Facebook Public Pages (PPs) based on relevant names such as politicians, firms, and NGOs. This process utilizes Facebook tools to gather high-frequency data on posts, including content, type, reactions, and views. Subsequently, content analysis techniques are applied to identify topics mentioned in posts and measure emotional intensity using methods like Emotion Lexicon and Contextual Sentiment Analysis.
Empirical analysis forms the crux of each work package's objectives. In the first part of DISRUPT, the impact of online campaigns on policy outcomes is assessed, combining the richness of various contexts. Notably, DISRUPT proposes to investigate the US context to document how online campaigns shape regulations, investigating U.S. Federal Regulation data, financial information, and advanced text analysis techniques. In addition, DISRUPT explores a distinctive feature of French regulation on mass layoff, by imposing a negotiation phase before the implementation of mass layoff plans. This offers unique opportunities to identify the effect of social media coverage on the outcome of the bargaining process, utilizing detailed text data on the mass layoff plans and information on the coverage of the negotiations both on French television and Facebook.
In the second part, DISRUPT investigates the complex interplay between social media and politics. Combining innovative data collection to characterize the presence, activity and engagement of politicians on Facebook, this second block of projects aim at analyzing which type of politicians benefit and loose from the penetration of social media in our societies, both in western democracies (with an emphasis on France), and in the Global South (with a focus on sub-Saharan Africa). This part of the DISRUPT project aims at exploring the richness of Facebook and electoral data in these two very different contexts, making use of frontline empirical methods to detect causal effects, and document relevant mechanisms.

Overall, the DISRUPT project aims to contribute significantly to the understanding of the complex interplay between social media, interest groups, collective action, electoral politics, and regional development in advanced democracies and developing countries. Through robust data acquisition, innovative online campaign data collection and analysis, and rigorous empirical analysis, the project endeavors to provide valuable insights into contemporary socio-political dynamics.

DISRUPT promises to make an immediate impact on the ongoing public discourse by shedding light on the role of social media in contemporary democracies, both advanced and developing. Its strength lies in its ability to offer robust empirical evidence regarding the potential benefits and drawbacks of social media in society, particularly focusing on who gains influence. Understanding this dynamic allows for further exploration into how social media shapes policy and local economic development. These twin objectives form the core of DISRUPT, situated at the intersection of diverse disciplines such as economics (including political economy, media economics, labor, and development economics) and other social sciences like sociology, communication, political marketing, and political science. Moreover, the project's data collection and analysis leverage recent advancements in computer science.

The DISRUPT team intends to produce 8 to 10 research papers to be disseminated first in working paper series and in SSRN/HAL, and later in international peer-reviewed journals, targeting leading general audience and top field journals. The team intends also to publish some of its work to top scientific journals for a general audience, including outlets such as Science or Nature when appropriate. The data pipeline (scripts, input & output datasets in open-access) created within the frame of the DISRUPT project will be posted on a dedicated webpage, following the FAIR Data Principles to allow replicability and foster the diffusion of DISRUPT’s technical output within the scientific community.

Social media raises the possibility of profound social, political, and economic changes, mobilizing large communities to push for disruptive moves. By allowing mass information diffusion at lower cost, social media offers an unprecedented platform to a broad number of agents in the society: from resource-poor grassroots organizations and marginalized politicians to private interest groups and powerful mainstream political parties. In this context, the critical question of who, in the society, is gaining from this major structural transformation in the market for information and influence, calls for deeper empirical investigations. In light of recent public debates about the role of social media in today’s society, DISRUPT proposes to carefully document the use of these platforms by politicians, non-profits, firms, workers and labor unions in advanced and developing economies, and address the multifaceted effects of social media on a rich set of political and economic outcomes (including regulation, employment and regional economic growth). To this aim, DISRUPT mobilizes innovative data collection technics with frontline econometric methodologies to provide a comprehensive empirical analysis addressing the following questions: How does social media affect influence across interest groups with unequal financial capacity? With which effects on the economy? Which politicians use social media in advanced and developing democracies? With which effects on politics, policy and regional economic development?