联系我们

论文查询结果	返回搜索

论文编号：	3915
作者编号：	1120080705
上传时间：	2011/12/6 12:59:12
中文题目：	在线评论的客户参与动机与评论有效性研究——基于中国大型B2C电子商务网站的实证分析
英文题目：	Customer Participation Motivation and the Helpfulness of Online Review: Data From Chinese B2C Electronic Commerce Websites
指导老师：	严建援
中文关键字：	在线评论，参与动机，有效性，文本挖掘，产品类型
英文关键字：	Online Review, Motivation, Helpfulness, Text Mining, Product Type
中文摘要：	近年，随着经济和社会的快速发展，电子商务已成为重要的零售业态，网络购物也成为人们购物的重要方式。电子商务和网络购物与实体销售最大的区别是商品只能通过网络可呈现的方式（文字、声音、图像等）达到消费终端即客户，商品的直观感知性和试用性受到极大限制。在线客户评论作为线下口碑的网络延伸形式，能够有效弥补上述不足。对电子商务销售商而言，在线客户评论有助于其实现网络营销、建立声誉系统和声誉传播的目的，同时对消费者而言，也有助于其做出正确的购买决策，降低网络购物的不确定性和风险。销售商希望获得更多更有效的在线客户评论，而消费者希望看到更有效对自己购买决策有帮助的其他客户发表的评论。因此，研究在线客户评论的来源和有效性影响因素具有重要的现实意义。本文基于传播学中的传播过程理论和霍夫兰德传播说服理论，站在中立的立场，研究客户参与在线评论的动机和在线评论的有效性影响因素，以三个大型中国B2C电子商务网站的5710条在线客户评论为对象，运用序贯互惠博弈分析和Tobit回归模型，研究“谁、说什么以及有什么效果”，目的是分别解决或应对网络购物和在线评论时代的“信息不足”和“信息过载”的问题。本研究的主要工作具体体现在以下三个方面：第一，对比分析了在线客户评论与传统口碑的异同，归纳总结出中国B2C电子商务网站上在线评论的特点，基于传播过程由传播者、传播内容、传播渠道、受传者和传播效果5个要素和环节组成的“5W模式”，建立了包括评论发表者、评论本体、接收者和电子商务网站四要素的在线客户评论理论分析框架。第二，归纳分析了知识共享、虚拟社区参与和口碑传播的动机，总结出包括自我导向性、社区导向性、商家导向性三个维度的九个在线客户评论参与动机。采取一对一深度访谈的方式，验证上述动机的同时发现了前人研究中尚未提到的“互惠”动机，然后运用序贯互惠博弈模型分析了客户参与发表在线评论的互惠心理和行为特点。第三，融合传播学、心理学、行为学和信息学等理论视角，基于归因理论、精细处理可能性模型、认知匹配理论、信息可获得性及诊断性理论，结合词性标注和词频统计等文本挖掘方法，提出影响在线客户评论有效性的四个维度：信息内容、信息来源、产品类型和发表时间。建立了在线客户评论有效性影响因素的模型，通过对中国大型B2C网站上评论数据的实证分析，对模型进行检验。根据结论，提出可供电子商务网站和客户参考的改进在线评论系统的措施。本文的主要研究结论和创新点如下：第一，针对以往从客户单方面、静态研究评论发表动机的不足，通过访谈发现了客户参与发表在线评论的“互惠”动机，并运用序贯博弈方法剖析了“客户与客户”双方互惠动机的动态作用机制和作用过程，发现信息共享、互惠倾向和相互信任在其中的重要作用。完善了在线评论参与动机体系，弥补了仅通过实验或实证方法研究参与动机问题的理论视角。访谈发现客户参与发表在线评论不仅局限于“自利”或“利他”的行为，还表现出“互惠”的心理。从博弈论的角度将发表评论的囚徒困境模型进行长期、动态和基于互惠心理的扩展，发现个人互惠动机较强，并且信任他人的情况下，利于评论的发表。第二，针对以往研究对评论文本内容缺乏深入分析的不足，运用文本挖掘技术客观地提取出在线评论文本内容的关键要素即语气强度、表述客观性和信息完整性，并构建了包括信息内容、信息来源、产品类型和发表时间四个维度的完整的在线客户评论有效性影响因素模型。运用文本挖掘和内容分析的方法，提取在线客户评论内容的关键属性，与以往研究相对比，为提炼在线评论有效性影响因素、建立更合理的模型提供了更可靠的依据。四维度的在线评论有效性影响因素模型改进并充实完善了其他学者提出的相关理论，模型得到来自中国三个大型B2C电子商务网站16种产品的5710条评论数据的验证，说明该模型具有合理性和普适性。第三，首次发现了评论深度与评论有效性的倒U型关系，结合对文本内容关键要素的实证分析，融合并解释了以往研究结论不一致的冲突，发现了评论语气强度、表述客观性与评论有效性的正相关关系，而信息完整性与评论有效性则成负相关关系。第四，针对以往对评论有效性的研究只考虑信息内容的不足，通过实证分析，首次发现并验证了评论来源与在线评论有效性的相关关系，同时验证了产品类型对评论深度、评论者情感极端性、其他评论者回应与评论有效性关系的调节作用。
英文摘要：	With the rapid development of information technology, world economy and society, e-commerce has gradually become an important retailing pattern, online shopping has also become a major way of shopping. One of the main differences between e-commerce and online shopping and that of the entity store is that commodity can reach the terminal customer only through the ways that can be rendered by the internet (such as text, voice, image and so on), as a result, visual perception and trial is limited. As a network extending form of word-of-mouth, online review can effectively compensate for these shortcomings. For vendors of e-commerce, online review helps them to achieve the purpose of network marketing, establishment of a reputation system and spread of reputation. For customers, it contributes to their decision making, reducing the uncertainty and risk of online shopping. Vendor wants to obtain more effective online reviews, and consumers would like to see more effective reviews to help their purchasing decisions. Therefore, it is of important practical significance to study the origin and influence factors of online customer reviews. Based on the diffusion theory and Hovland communication and persuasion theory, this paper studied the motivation for customers to participate in online reviews and factors influencing the helpfulness of customer review, standing in the neutral position. Thesis took 5710 online customer reviews of Chinese B2C electronic commerce websites as our object of study, and studied “who, what to say and helpfulness”, and aim to resolved the problems of "insufficient information" and “information overload” in the online shopping era. The main works of our research are as follows: First, compared the similarities and differences between online reviews and traditional word-of-mouth, and got the characteristics of online reviews of Chinese B2C e-commerce website. Thesis built the theoretical analysis framework of online reviews, which including reviewer, contents of the review, receivers and the e-commerce website, based on the “5W Mode” of who, says what, in which channel, to whom, and with what effect. Second, analyzed the motivations of knowledge sharing, participation in virtual community and word-of-mouth diffusion, and summed up with nine participation motivations of online reviewers, from three dimensions of self-directed, community oriented, and business oriented. Thesis discovered the “reciprocity” motivation while verifying the above mentioned motivation, with one-to-one in-depth interview. Then, Thesis analyzed the mutual reciprocal psychology and behavior characteristics when customers publish reviews, with sequential reciprocity game model. Third, based on the theories of communication, psychology, behavior and information science, thesis proposed the four dimensions that affect the helpfulness of customers’ reviews, which are information content, information sources, product type and other dimension, from the attribute theory, elaboration likelihood model, cognitive fit theory, information assessibility and diagnosticity theory, with the text mining methods of part-of-speech tagging and word frequency statistics. Established the model of influence factors of the validity of online customer reviews, and validate the model with empirical data of 5710 reviews of Chinese large B2C websites. According to the conclusions, Thesis provided some measures for e-commerce sites and customers to improve online review system. The main research conclusion and innovations are as follows: First, for the shortage of unilateral and static research of participation motivation, thesis discovered the customers’ “reciprocity” motivation to publish reviews through interview. Then using sequential game theory analyzed “customer and customer” mutual reciprocity motivation and dynamic mechanism. Found that information sharing, reciprocity propensity and mutual trust play an important role. Those foundations make up the theoretical perspectives of participation motivation that only from experiment and empirical analysis. Thesis found that customer participate in the online reviews not only for “egoism” or “altruism”, they also show “mutual reciprocity” tendency. Thesis extended the prisoner’s dilemma model that analyze the online reviews from the angle of long-term, dynamic and mutual reciprocal psychological, and found that when individuals have stronger reciprocity motivations, and trust others, they would more likely to publish reviews. Second, for the shortage of analysis of reviews’ text content, using data mining technology objectively thesis extracted the key attributes of online reviews include tone intensity, representations of objectivity, and information integrity. Thesis built a comprehensive model that measures the helpfulness of reviews, based on information content, sources of information, product type and the others. Using text miming and content analysis methodology thesis provided a more reliable basis for the establishment of a more reasonable model, and refining the factors affect the helpfulness of online review. The reviews’ helpfulness model improves the relative theories that analyze the helpfulness of reviews. The empirical data are from online e-commerce websites in China, and using the objective data from their websites, these can greatly avoid the error that questionnaire may cause. Third, thesis found that there is an inverted U shape relationship among the depth and emotion of reviewers for the first time. Based on the empirical analysis of reviews contents, thesis integrated and explained the previous inconsistent findings and conflicts. Thesis found positive correlation between tone intensity, objectivity of reviews with the helpfulness of reviews, but negative correlation between information integrity with reviews’ helpfulness. Fourth, for the shortage of only considering the information content, through empirical analysis, thesis firstly found and examined the correlation between sources of information and online reviews’ helpfulness. Product type really played a partial moderating role in some of the above mentioned relationships.
查看全文：	预览  下载(下载需要进行登录)