The dynamics of online word-of-mouth and product sales—An empirical investigation of the movie industry
Introduction
Word-of-mouth (WOM) has been recognized as one of the most influential resources of information transmission since the beginning of human society (Godes and Mayzlin 2004; Maxham and Netemeyer 2002; Reynolds and Beatty 1999). However, conventional interpersonal WOM communication is only effective within limited social contact boundaries, and the influence diminishes quickly over time and distance (Bhatnagar and Ghose 2004; Ellison and Fudenberg 1995). The advances of information technology and the emergence of online social network sites have profoundly changed the way information is transmitted and have transcended the traditional limitations of WOM (Laroche et al. 2005). The otherwise fleeting WOM targeted to one or a few friends has been transformed into enduring messages visible to the entire world. As a result, online WOM plays an increasingly significant role in consumer purchase decisions.
Online WOM presents both challenges and opportunities to retailers. On the one hand ,WOM provides an alternative source of information to consumers, thus reducing retailersrsquo; ability to influence these consumers through traditional marketing and advertising channels. Prior studies show that a variety of aspects of WOM influence retail sales. Some found that WOM dispersion (Godes and Mayzlin 2004) and valence (Chevalier and Mayzlin 2006; Forman, Ghose, andWiesenfeld 2008) have significant effects on product sales, while others found that WOM volume serves as the key driver of product sales (Chen, Wu, and Yoon 2004; Liu 2006). On the other hand, online WOM provides a new venue for retailers to reach consumers and to strategically influence consumer opinions. Anecdotal evidence has surfaced in recent years suggesting that online WOM could be successfully leveraged as a new marketing tool (Dellarocas2003). A unique aspect of the WOM effect that distinguishes it from more traditional marketing effects is the positive feedback mechanism between WOM and product sales. That is, WOM leads to more product sales, which in turn generate more WOM and then more product sales. The positive feedback mechanism indicates that WOM is not only a driving force in consumer purchase but also an outcome of retail sales ( Godes and Mayzlin 2004; Srinivasan, Anderson, and Ponnavolu 2002). Prior studies on WOM have not fully recognized this unique nature of WOM effect and often treat WOM as exogenous, like traditional marketing effects (Chen et al. 2004; Liu 2006). Ignoring WOMs dual roles of precursor and outcome may misplace causality and lead to erroneous results. The objectives of this study, therefore, are to explicitly model the positive feedback mechanism between WOM and retail sales and identify their dynamic interrelationship. We propose a simultaneous equation system to fully capture the dual nature of online WOM and its dynamic evolution
in a panel data setting.
We have chosen the movie industry as our research context because industry experts agree that WOM is a critical factor underlying a moviersquo;s staying power, which leads to its ultimate financial success (Elberse and Eliashberg 2003). In addition, the movie industry has by far received the most attention in marketing literature on WOM, which allows in-depth comparison of our results with those of previous studies. We, however, note that movies are a unique type of experience goods and the results from the industry do not necessarily generalize to other retailing sectors. Rather, our goal is to use the movie industry as a context to highlight the importance of considering the dynamics of and the interrelationship between retail sales and online WOM and to demonstrate the validity of the simultaneous equation approach in this setting. We found that both a moviersquo;s box office revenue and WOM valence significantly influence WOM volume. WOM volume in turn leads to higher box , office performance. Our results clarify conflicting results reported in earlier studies with regard the influence of user ratings on box office revenue. We show that user ratings do not directly influence box office revenue. However, they affect box office revenue indirectly through WOM volume.
Online WOM in the movie industry takes many forms, including online reviews, discussion boards, chat rooms, blogs, wikis, and others. In this study, we focus on online user reviews because statistics suggest that user reviews are more prevalent than other forms of WOM communication in the movie industry. Beyond volume, another subtle but important difference between online user reviews and other types of WOM is that user reviews usually reflect user experience and consumer satisfaction, which are mainly viewed as a source of product information (Chen and Xie 2004; Li and Hitt 2008). Meanwhile, other types of WOM, such as discussions in online community sites, reflect more about consumer expectation, which could be heavily influenced by social structure (Gopal et al. 2006; Liu 2006).
The rest of the paper is organized as follows. The next section provides the literature review followed by the discussion of our conceptual framework and research hypo theses. We then describe our sources of data and the empirical model and estimation. Main findings are presented and discussed next, and the paper ends with a discussion of implications, limitations, and future research.
Empirical model specification
The development of our empirical model is guided by the following considerations. First, as we are interested in the drivers of both box office revenue and WOM, we construct a system of two interdependent equations: one equation with daily revenue as the dependent variable (the revenue equation) and the other
with WOM volume as the dependent variable (the WOM equation). We assume that in ea
剩余内容已隐藏，支付完成后下载完整资料
附录A 译文
在线口碑和产品销售动态——对电影行业的实证调查
介绍
口碑自人类社会开始以来，已经被公认为最有影响力的信息传递的资源(Godes and Mayzlin 2004; Maxham and Netemeyer 2002; Reynolds and Beatty 1999).然而，传统的人际间口碑交流只有在有限的社会接触范围内才会起作用，其影响随时间和空间的推移迅速减弱(Bhatnagar and Ghose 2004; Ellison and Fudenberg 1995)。信息技术的进步和在线社交网站的出现，深刻地改变了信息传播的方式，并已超过了传统口碑的界限(Laroche et al. 2005)。另外，在一个或几个朋友间传播的短暂口碑，已经被看作是能透过此看见世界的信息。因此，在线口碑在消费者购买决策中发挥着日益重要的作用。
在线口碑对零售商来说，既是机遇也是挑战。一方面，口碑为消费者提供了可供选择的信息源，从而降低了零售商通过传统市场营销和广告渠道影响消费者的能力。先前的研究表明，口碑的各个方面会影响到零售的销量。一些研究发现，口碑量(Godes and Mayzlin 2004)和价数(Chevalier and Mayzlin 2006; Forman, Ghose, andWiesenfeld 2008)对产品销量产生重大影响，而另一些研究发现，口碑量被看作是产品销量的主要驱动力(Chen, Wu, and Yoon 2004; Liu 2006).。另一方面，在线口碑为零售商提供一个接触到消费者和战略影响消费者意见的场所。近几年出现的传闻证据表明，在线口碑已作为新的营销工具被成功应用(Dellarocas2003).。不同于传统的市场营销效果，口碑效应的独特之处在于口碑和产品销量的积极反馈进程，也就是说，口碑会提高产品销量，反过来产生更多的口碑和更多的产品销量。积极的反馈机制表明，口碑不仅是消费者购买的驱动力，而且是零售销量的结果( Godes and Mayzlin 2004; Srinivasan, Anderson, and Ponnavolu 2002)。此前对口碑的研究没有充分认识到口碑效果的独特性，而是把口碑看成是像传统营销效果一样的外因(Chen et al. 2004; Liu 2006)，忽略口碑的双重先导角色和可能因误置角色而导致错误结果的影响。因此，这项研究的目的是要明确口碑和产品销量模式的积极反馈作用，并确定他们之间的动态关系。我们提出了一个联立方程系统，以充分诠释在线口碑的双重性质以及它在数据库设置中的动态演进。
我们选择了电影产业作为研究范畴。因为业界专家一致认为，口碑是能令电影持续发挥其功效的一个关键因素，从而实现最终财务上的成功(Elberse and Eliashberg 2003)。此外，电影界迄今为止在文学营销口碑上受到重视，这使得我们的成果能和之前的研究进行深入的比较。然而，我们注意到电影作为一种特殊的体验产品类型，其产业的结果并不一定能够推广到其他零售行业。相反，我们的目标是利用电影产业作为一个内容来强调考虑零售销量和在线口碑的动态和他们之间关系的重要性，并说明设置的联立方程方法的正确性。我们发现，一部电影的票房收入和口碑效价明显地影响着口碑量，口碑量反过来会带来更高的票房和更好的成绩。我们的研究结果澄清了早期研究中用户评分对票房收入影响的博弈。我们表明，用户评分不会直接影响票房收入。但是，他们会通过口碑间接影响到票房收入。
此外，我们的研究还证实了在线口碑不仅是先导，而且是产品销量的结果。我们表明， 忽视口碑的双重性质会导致错误的结果。
电影产业的在线口碑有多种表现形式，包括在线评论，讨论版，聊天室，博客，维基和其他。在此研究中，我们侧重于在线用户评论，因为数据显示，在电影产业中，在线用户评论比其他口碑交流更加普遍。除了量之外，在线用户评论和其他形式的口碑之间的重要微妙区别在于，用户评论通常反映了用户的经历和满意度，这被看作是产品的信息源之一(Chen and Xie 2004; Li and Hitt 2008).。与此同时，口碑的其他类型，比如较能反映消费者的预期的在线社区网站的讨论，深受社会结构的影响(Gopal et al. 2006; Liu 2006)。
本文的其余部分组织如下。下一节将提供支撑我们概念框架和研究模型的文献。然后，我们描述我们的数据来源和经验模型及其预测。之后，将描述和讨论主要的调查结果，以及讨论的意义、限制及未来的研究。
实证模型详述
我们的实证模型发展遵循以下考虑。首先，由于我们对电影票房收入和口碑驱动力感兴趣，我们构建了两个相互依存的系统，一个是每日收入作为因变量方程（收入方程），另一个是口碑作为因变量的方程（口碑方程）。我们假设在每个阶段（例如，一天），两个方程的错误是相关的，这意味着不包括在我们模型的因素会同时影响着电影的收入和口碑。
其次，认识到消费者对电影的选择行为和口碑之间相互作用会超越同期(Elberse and Eliashberg 2003)，我们发展了一个动态方程系统。在收入方程中，不仅包括同期的每日口碑量，而且还有滞后的因素。同样，在口碑方程中，滞后的收入因素也同样存在着。这样的规范有助于识别两个联立方程系统，因为滞后因素属于每个方程的外在变量。此外，继现存的研究，我们在模型中使用线性方程(e.g., Elberse and Eliashberg 2003; Liu2006)。线性方程符合消费者决策过程的各个阶段，一部电影的销量可以看作是应用到消费群的一系列条件概率。一个数可转换成经验估计的线性关系模型，此外，线性方程中回归变量的平滑分布，以及线性形式的估计系数可直接反应非独立和独立变量的弹性。第三，为了控制更多可影响影响电影收入和口碑的特有因素，如预算、市场营销、明星和其他(Basuroy et al. 2003; Elberse and Eliashberg 2003; Liu 2006)，我们把在模型中通过增加特定虚拟变量所带来的固定效果包括在内。固定效果捕获了任何变化因素，包括内在的电影特色，评论家评论和其他外界性因素。另外，固定效应估计允许与其他变量相关的任意误差，从而使估计结果更加稳健。
意义、限制和未来研究
我们的模型详细说明了双重因果关系，并揭示了在线口碑和产品销量之间的积极反馈机制。我们的研究有力地支持了考虑口碑的内在性质和消费者消费行为之间的相互依存关系。从3SLS（更强大的统计方法）和OLS的明显不同得出，利用简单回归技术的现存研究可能会得出关于口碑效应方向和大小的偏见结论。我们的结果验证了我们关于在线用户评论数量和零售销量双赢关系的断言。
我们的研究也是对先前关于口碑量、口碑效价、票房销量之间关系研究(Basuroy et al. 2003; Eliashberg and Shugan 1997; Liu 2006)的重要延伸。先前的研究一直注重口碑量的直接影响和票房收入价的效价，并发现主要的解释力量来自于口碑量而不是口碑效价(Liu 2006)。我们的研究拓展了考虑口碑量和口碑效价相互作用的方法，我发现，虽然口碑效价不能直接影响收入，但更高的口碑效价能通过产生更多的口碑量间接提高票房收入。此研究对口碑文献的贡献是多方面，从方法论的角度来看，我们揭示了分离先导和销售结果的口碑效应的重要性。同时，我们的结果强调了，在数字环境中，使用动态系统和高频数据对研究口碑效应的重要性。从管理角度来看，我们说明了口碑效价和口碑量在影响产品销量中扮演着不同的角色。我们还说明，口碑效价时间序列的变化影响着能带来更高销量的口碑量。我们的研究支持在线口碑进程对销量有重要影响的想法，这表明企业应支持和丰富口碑活动。
有很多机会去延伸现有的研究，我们研究中的一项重要和有趣的扩展是去调查消费者在口碑信息，尤其是数字环境中的决策过程。此外，并非所有的口碑都是平等的，消费者需要从网上所有反馈和建议中寻找到真实和诚实的意见。在这种情况下，消费者如何从信息源中找到可信赖的信息，是今后研究中特别有趣的。
在线用户评论只是一种，连接消费者的媒介。作为一项新媒介的生成和分布的在线社区(e.g., www.YouTube.com, www.Flickr.com, and www.Digg.com)受欢迎程度的急剧增长，在互联网上引起了新的关注。与在线网站评论不同，我们在这里探讨，在线社交社区鼓励用户之间的互动，这可能会改变口碑的动态分布。因此，本次研究中使用的模型方法可能不能充分地反映所研究的内容。对新媒介中在线口碑效应的描述和确定的新研究，有利于我们理解在线连接消费者的媒介对市场营销和零售战略所产生的影响。
我们的分析，根据需要对选择发帖评论和在雅虎评论电影的在线用户进行限制。因此，我们的研究是对这样一个有条件的用户群的估计。虽然这样的限制不一定有偏于面板数据的预测结果，他们应该被解释为运用到一个在线用户的自我选择设置。此外，本为只是对张贴发布口碑和销量之间关系的研究。不过，口碑在电影发布前是肯定存在的，电影制片厂作出各种营销努力以促进口碑(Liu 2006)。关于雅虎电影的一项详细调查显示了预发布口碑和发布口碑之间重要和有趣的区别。我们注意到，在雅虎电影讨论版的预发布口碑活动中心，帖子主要反映了消费者期望。同时，在雅虎电影用户评论网站的发口碑贴中心，帖子主要反映消费者的满意度和产品体验。这种差别表明了在预发布口碑和发布口碑之间，确实存在着不同的机制。因此，对现有研究的一项重要延伸应研究预发布和发布在线口碑之间的动态关系并区分它们的影响。
附录B 外文原文
The dynamics of online word-of-mouth and product sales—An empirical investigation of the movie industry
Introduction
Word-of-mouth (WOM) has been recognized as one of the most influential resources of information transmission since the beginning of human society (Godes and Mayzlin 2004; Maxham and Netemeyer 2002; Reynolds and Beatty 1999). However, conventional interpersonal WOM communication is only effective within limited social contact boundaries, and the influence diminishes quickly over time and distance (Bhatnagar and Ghose 2004; Ellison and Fudenberg 1995). The advances of information technology and the emergence of online social network sites have profoundly changed the way information is transmitted and have transcended the traditional limitations of WOM (Laroche et al. 2005). The otherwise fleeting WOM targeted to one or a few friends has been transformed into enduring messages visible to the entire world. As a result, online WOM plays an increasingly significant role in consumer purchase decisions.
Online WOM presents both challenges and opportunities to retailers. On the one hand ,WOM provides an alternative source of information to consumers, thus reducing retailersrsquo; ability to influence these consumers through traditional marketing and advertising channels. Prior studies show that a variety of aspects of WOM influence retail sales. Some found that WOM dispersion (Godes and Mayzlin 2004) and valence (Chevalier and Mayzlin 2006; Forman, Ghose, andWiesenfeld 2008) have significant effects on product sales, while others found that WOM volume serves as the key driver of product sales (Chen, Wu, and Yoon 2004; Liu 2006). On the other hand, online WOM provides
剩余内容已隐藏，支付完成后下载完整资料
资料编号：[481776]，资料为PDF文档或Word文档，PDF文档可免费转换为Word
以上是毕业论文外文翻译，课题毕业论文、任务书、文献综述、开题报告、程序设计、图纸设计等资料可联系客服协助查找。
