A replication and extension of the Personal Social Media Ecosystem Framework

Abstract

The recently forwarded Personal Social Media Ecosystem Framework (PSMEF) allows researchers to study social media in terms of generalized types of user interfaces. This study formally extended the PSMEF via the Digital User Interface Model and replicated previous work by evidencing the existence of new (e.g., Overtly Algorithmic Content Pages) and validating previously identified types of user interfaces (e.g., Home Pages and Chats/Messages) that make up individuals' personal social media environments. Using topic modeling (i.e., Latent Dirichlet Allocation) and a novel mixed methods approach (i.e., schematic semantic network analysis), we quantitatively evidenced four distinct classes of user interfaces based on open-ended descriptions that participants provided for six popular social media platforms (i.e., Instagram, Snapchat, Facebook, Twitter, TikTok, and YouTube). Results inform on the qualitative differences between distinct user interface classes that underwrite users' experiences over social media, with implications for conceptualization and operationalization related to social media use.