Purpose The purpose of this study is to integrate network externality and service quality as antecedents to the synthesis of expectation-confirmation model (ECM) and perceived ease of use (PEOU) in explaining customers' continuance intention of internet-based sharing economy service platforms. Design/methodology/approach Sample data for this study were collected from students enrolled in a comprehensive university in Taipei, Taiwan. A total of 600 questionnaires were distributed in various campus locations, and 510 usable questionnaires were analyzed in this study, with a usable response rate of 85.0%. Collected data were analyzed using structural equation modeling. Findings The primary findings are as follows. First, service enablers' online service quality positively affected customers' perceived usefulness (PU), confirmation and PEOU of internet-based sharing economy service platforms, which directly or indirectly explained customers' satisfaction and continuance intention of the platforms. Next, service providers' offline service quality positively influenced customers' PU of internet-based sharing economy service platforms, which in turn caused customers' satisfaction and continuance intention of the platforms. Finally, network externality from customers' side and service providers' side all positively affected customers' PU of internet-based sharing economy service platforms, which in turn led to customers' continuance intention of the platforms. Originality/value First, this study contributes to the identification of network externality and service quality that may reveal deep insights to the understanding of customers' continuance intention of internet-based sharing economy service platforms greatly driven by their confirmation of expectations and satisfaction with the platforms. Next, the empirical evidence on capturing ECM and PEOU for completely explaining the antecedents of customers' continuance intention of internet-based sharing economy service platforms is well documented. Hence, this study's findings have significantly shed light on the possible formulation of a richer post-adoption model.
