消费者愿意采纳推荐吗?--基于信息系统成功-技术接受模型

Are Consumers Willing to Adopt Recommendations?Based on Information Systems Success-Technology Acceptance Model

消费者愿意采纳推荐吗?这是网商成功应用推荐系统的关键。基于技术接受模型,现有研究发现:感知有用性与感知易用性共同影响推荐采纳意向。但哪些与推荐系统相关的因素影响感知有用性与感知易用性?现有研究分析不全面,缺乏理论基础。基于此,笔者结合信息系统成功模型与技术接受模型,构建了购物网站服务质量、推荐系统质量和推荐信息质量对感知有用性、感知易用性和推荐采纳意向的影响模型。使用问卷调查法收集数据,运用结构方程模型分析数据。研究结果表明:购物网站服务质量、推荐系统质量和推荐信息质量通过中介变量感知有用性与感知易用性最终影响推荐采纳意向;推荐信息质量对推荐采纳意向的总效应最强,其次为购物网站服务质量,最弱的是推荐系统质量。此研究成果推进了推荐系统在营销领域的研究进展,拓展了信息系统成功模型与技术接受模型在网络购物环境下的应用,并对网商改善其推荐系统有指导意义。

Are consumers willing to adopt recommendations？ This is the key to success in application of recommender systems. Based on the technology acceptance model （TAM）, the existing studies have found that both perceived usefulness and perceived ease of use influence adoption intention of recommendations. However, what factors related to recommender systems influence perceived usefulness and perceived ease of use？ The ana- lysis of existing studies is not comprehensive, and lacks theoretical basis. Based on this, we combined the TAM and the information systems success model （ISSM） to develop a model to elaluate website service quality, re- commender system quality, and recommendation information quality affecting perceived usefulness, perceived ease of use, and adoption intention of recommendations. We collected data by means of questionnaires, and conducted data analysis with structural equation modeling （SEM）. Our results indicate that website service qua- lity, recommender system quality, and recommendation information quality ultimately influence adoption inten- tion of recommendations via the mediating constructs perceived usefulness and perceived ease of use; recom- mendation information quality has the strongest total effect on adoption intention of recommendations, followed by website service quality and recommender system quality, respectively. These research results boost research progress on recommender system in marketing, extend the application of the ISSM and the TAM in the online shopping environment, and provide managerial implications for e-shops to improve their recommender systems.