关键词: 共享短租平台, 相似吸引理论, 感知相似性, 顾客持续参与行为

Abstract: With the development of information and media technology, the sharing economy, especially the peer-to-peer (P2P)sustainable consumption model, is effectively transforming people's consumption lifestyle with its concepts of environmental friendliness and efficiency. Sharing accommodation platforms such as Airbnb, Xiaozhu, and Couchsurfing optimize the allocation of housing resources worldwide through internet platforms, offering individuals the opportunity to share their idle housing with strangers. However, as anyone can become a local host, factors like lack of stable expectations and specialized service training become sources of risk (e. g. , monetary, emotional, and personal safety risks)for customers choosing to share with a host. After consumers select online hosts on the platform for transactions, they need to communicate face-to-face with unfamiliar hosts offline. Given potential monetary and personal safety risks, customers are eager to filter out more reliable hosts through online platform information. Consequently, due to the relatively low institutional trust environment of sharing platforms, customers' continuous engagement behavior has remained a crucial issue to be addressed in the field of the sharing economy.
In the online transaction environment, information sources canrelease signals about the quality of products or services to attract information recipients. Among these signals, source similarity is considered a significant attraction signal that can effectively enhance communication between sources and recipients. Based on the similarity attraction theory, relationships develop as individuals come into contact, and personal attraction generated through similarity is more likely to stimulate the formation of interpersonal relationships, fostering “friendship and intimacy” characteristics between strangers. Therefore, this study analyzes the impact of perceived similarity among sharing economy platform hosts on customers' continuous engagement behavior, and explores the moderating effects of companionship and property type. To achieve this objective, the study proposes a triple-interaction model, and by obtaining web-scraped data from the Airbnb platform, establishes a mixed-effects negative binomial regression model to conduct hypothesis testing, so as to investigate the triple interaction effects of perceived similarity, companionship, and property type on customers' continuous engagement behavior. Findings of the research are as follows: Perceived similarity significantly and positively influences customers' continuous engagement behavior; Companionship negatively moderates the impact of perceived similarity on customers' continuous engagement behavior; The moderating effect of companionship on the relationship between perceived similarity and customers' continuous engagement behavior is influenced by property type, with a significantly enhanced negative moderating effect of companionship in shared room compared to independent rooms.
The theoretical contributions of this study are as follows: First, it advances new insights into the issues of information asymmetry and host-guest interaction in sharing accommodation platforms. Second, it examines the impact of perceived similarity, an important social attraction signal of information sources (hosts), on customers' continuous engagement behavior during the online purchasing decision phase, supplementing the applicability of similarity attraction theory within the context of the sharing economy platform. Third, it refines the boundary conditions of the influence of perceived similarity on customers' continuous participation behavior. By introducing companionship and property type as two moderating variables, a triple-interaction model is constructed, offering new research ideas and directions for future studies.

Key words: Sharing Accommodation Platforms, Similarity Attraction Theory, Perceived Similarity, Customers' Engagement Behavior