Social Paleontology on Twitter: A Case Study of Topic Archetypes, Network Composition, and Structure

Abstract

Social paleontology is a burgeoning field of research that seeks to understand the natural world through the collection, preparation, curation, and study of fossils via online communities. Such a community represents an ideal case for examining scientific practice as the expression of conversation topics in relation to the people who participate. Using Communities of Practice as a theoretical framework, we consider interactions within an egocentric Twitter network over a 397-day period to identify topic archetypes within the community, examine how such topic archetypes act as expressions of behavior that are indicative of community processes, and provide empirical evidence for detecting and indicating the health of an online community. Data were collected continuously and analyzed with a combination of topic modeling and social network analysis. Four unique archetypes were characterized based on the level of activity and longevity of interest. Participants for each were diverse, but not different. Structural differences in each network were noted with high levels of inter-group information flow within certain archetypes. Archetypes were interpreted using the life cycle states for Communities of Practice; sustained conversations and piques of interest indicate healthy online communities. These findings can inform efforts to design, implement, and research online, scientific communities.