The influence of information similarity overload on users' withdrawal behaviour from social communities on SNSs.

Excessive and repetitive information is a common experience in online communities. While prior research has focused on general information overload, the effects of similarity overload, where content is redundant rather than voluminous, remain less clear. Drawing on construal level theory, this study examines how similarity overload contributes to users’ withdrawal from social communities on social networking sites (SNSs). The model tests the mediating roles of boredom and perceived misfit, and the moderating influence of perceived social pressure. A face-to-face survey was conducted with 282 SNS community users in Vietnam. Established scales were adapted to measure similarity overload, boredom, misfit, perceived social pressure, and withdrawal behaviour. Partial least squares structural equation modelling (PLS-SEM) was applied, with bootstrapping to assess mediation and moderation. Reliability, validity and multicollinearity were confirmed. Results show that similarity overload is positively associated with withdrawal, but its effect is primarily indirect. Both boredom and perceived misfit significantly mediated the relationship, revealing affective and cognitive pathways. Perceived social pressure further intensified these effects, with users embedded in strong normative climates more likely to disengage under similarity overload. The findings identify similarity overload as a distinct dimension of information stress that reduces engagement by lowering intrinsic motivation and personal and community alignment. To this end, the study advances construal level theory in digital contexts and offers practical guidance for community managers to mitigate withdrawal by diversifying content and monitoring social norms.