CooRTweet: A Generalized R Software for Coordinated Network Detection

This paper introduces CooRTweet, an innovative R package designed for detecting and analyzing coordinated behavior. CooRTweet’s distinctiveness lies in its essential architecture, derived from a minimal definition of coordinated behavior that captures its core elements in an abstract way. This approach makes it possible for the tool to be applied to the widest range of cases, from mono-modal network analysis on a single social media platform, to multi-modal and cross-platform network analysis, and to any types of objects shared by a network, whether singular identical objects (e.g., the same tweet), similar objects (e.g., clusters of similar images), or complex objects (e.g., a combination of hashtags, images, and emojis). Additionally, it offers a comprehensive view of coordinated activities that include both explicit coordination and organic forms of content sharing. The comprehensive architecture of CooRTweet provides flexibility and a broad scope for analyzing coordinated activities across various digital landscapes. This positions it as a distinctive resource for researchers investigating coordinated communication online. More generally, CooRTweet provides a valuable example to methodologists and research tool developers of how software tools for research can be developed in a generalized and thus flexible way. This is particularly important for social media research, given how quickly new APIs are being released, modified, and even shut down. This paper aims to provide an introduction to CooRTweet and the analysis of coordinated behavior, demonstrating the software’s application through a case study of cross-platform coordinated behavior during the 2021 German elections.