The Influence Of Interaction On Consumer’s Purchase Intention In Online Community

Online community has become the new interactive platform for consumers.Consumers’ interactive behavior in online community, which is different from interactivebehavior in traditional context, has attracted much attention from both academic andindustrial fields. Virtual community and SNS are two main types of online community.There are remarkable differences between the two communities. Virtual community andSNS will coexist for a long time. Companies’ businesses involve how to promote in thesecommunities. Therefore, the aim of this paper is to explore the influence of interaction onconsumers’ purchase intention in the context of online communities, and compare thedifference of the influence between virtual community and SNS context.Based on the Technology Acceptance Model (TAM), this paper developed atheoretical model to predict the influence of interaction on consumers’ purchase intentionin online communities. The theoretical model took interaction dimensions as stimulus,perceived usefulness, cognitive trust and perceived expectation fit as psychologicalreaction, and purchase intention as behavioral reaction. Using data from online surveys,the proposed model was tested by using the Partial Least Squares (PLS) technique. Theprocess of data analyses contains four parts: two preliminary studies, a full model analysis(not containing moderator) and a moderator analysis. The contents and results of theseparts are as follow:First, the first preliminary study: the development of a plan-interaction-reactionmodel. Based on the TAM, this paper developed a theoretical model to predict thepsychological and behavioral reaction of consumers in the context of interaction in onlinecommunities, which is named plan-interaction-reaction model. Using data from onlinesurveys of consumers who have the experience of interaction in online communitiesbefore purchasing digital products/household appliance, the proposed model was tested byusing PLS. The results show that the plan-interaction-reaction model is well suitedto explain consumers’ reaction process of interaction in online communities: perceivedusefulness, cognitive trust and perceived expectation fit are antecedents of purchaseintention; perceived usefulness and cognitive trust have positive impact on perceived expectation fit; cognitive trust has a positive impact on perceived usefulness. In addition,the plan-interaction-reaction model explained a significant proportion of the variance inconsumers’ purchase intention, which is3%higher than TAM in the whole onlinecommunities context (R2=3%) and3%higher than TAM in the virtual communitiescontext (R2=3%) and4%higher than TAM in the SNS context (R2=4%).Second, the second preliminary study: the development of interaction dimensions.Based on the Information Communication Model, this paper developed the dimensionsand scale of interaction in online communities. Using data from online surveys ofconsumers who have the experience of interaction in online communities beforepurchasing products, the scale was tested by using PLS. The results show that consumers’interaction characteristics in online communities can be tested by using four dimensions,namely source quality, tie strength, information quality and interaction atmosphere.Third, the full proposed model (not containing moderator): the development of atheoretical model to predict the influence of interaction on consumers’ purchase intentionin online communities. Based on the conclusions of the two preliminary studies, usingdata from online surveys of consumers who have the experience of interaction in onlinecommunities before purchasing products, the full proposed model was tested by using PLS.The results show: perceived usefulness, cognitive trust and perceived expectation fit areantecedents of purchase intention; perceived usefulness and cognitive trust have positiveimpact on perceived expectation fit; cognitive trust, information quality and interactionatmosphere have positive impact on perceived usefulness; source quality, tie strength,information quality and interaction atmosphere have positive impact on cognitive trust.Fourth, the analysis of moderator: a comparison between virtual community and SNS.By using data collected in the third study, the full proposed model was tested in thecontext of virtual community and SNS context by using PLS. Then, multi-groups analysiswas explored to compare the difference between the two groups. The results show: sourcequality has significant effect on perceived usefulness in virtual community context, buthas none in SNS context; tie strength has no significant effect on perceived usefulness inboth communities; information quality and interaction atmosphere have significant effecton perceived usefulness in virtual community context, but has none in SNS context;source quality has significant effect on cognitive trust in both communities, but more stronger in SNS context; tie strength and information quality have significant effect oncognitive trust in virtual community context, but has none in SNS context; interactionatmosphere has significant effect on cognitive trust in both communities.As last, some marketing implications were developed based on the conclusion of theempirical studies.