外文翻譯--在線口碑和產(chǎn)品銷售動(dòng)態(tài)——對(duì)電影行業(yè)的實(shí)證調(diào)查

1、<p><b>　　中文3310字</b></p><p><b>　　畢業(yè)論文外文翻譯</b></p><p><b>　　一、外文原文</b></p><p>　　標(biāo)題：The dynamics of online word-of-mouth and product sales—An e

2、mpirical investigation of the movie industry</p><p><b>　　原文：</b></p><p>　　Introduction</p><p>　　Word-of-mouth (WOM) has been recognized as one of the most influential re

3、sources 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 effect

4、ive 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 th</p>&

5、lt;p>　　Online WOM presents both challenges and opportunities to retailers. On the one hand ,WOM provides an alternative source of information to consumers, thus reducing retailers’ ability to influence these consumers

6、 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 Mayz

7、lin 2006; Forman, Ghose, andWiesenfeld 2008) have significant effects </p><p>　　We have chosen the movie industry as our research context because industry experts agree that WOM is a critical factor underlyi

8、ng a movie’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

9、 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 f</p><p>　　Online WOM in the movie industry takes many

10、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

11、 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 satisf

12、action, which ar</p><p>　　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 th

13、en 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.</p>&

14、lt;p>　　Empirical model specification</p><p>　　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 WO

15、M, we construct a system of two interdependent equations: one equation with daily revenue as the dependent variable (the revenue equation) and the other</p><p>　　with WOM volume as the dependent variable (th

16、e WOM equation). We assume that in each time period (i.e., day), the errors in the two equations may be correlated, which implies that factors not included in our model could simultaneously influence both movie revenue a

17、nd WOM.</p><p>　　Second, recognizing that interactions between consumers’ movie-going behavior and WOM can go beyond the concurrent term (Elberse and Eliashberg 2003), we develop a system of dynamic equation

18、s. That is, in the revenue equation, we include</p><p>　　not only the contemporaneous term of daily WOM volume, but also multi-lag terms. Likewise, in the WOM equation, multi-lag revenue terms are also incor

19、porated. Such a specification also helps identify both equations for the simultaneous equation system since the lagged terms are exogenous variables in either equation. In addition, following extant research, we use alog

20、-linear formulation (e.g., Elberse and Eliashberg 2003; Liu2006) in our model. The log-linear formulation is consistent with</p><p>　　theoretical models of a multistage consumer decision-making process, wher

21、e sales of a movie can be viewed as a series of conditional probabilities applied to the consumer base. A log transformation converts the relationship into a linear form for empirical estimation. Moreover, log transforma

22、tion smoothes the distribution of variables in the linear regression, and the estimated coefficients of the log-linear form directly reflect the elasticity of independent and dependent variables. Third, to c</p>&

23、lt;p>　　fixed effects estimation also allows the error term to be arbitrarily correlated with other explanatory variables, thus making the estimation results more robust.</p><p>　　Implications, limitations

24、, and future research </p><p>　　Our model specifies the dual causal relationship and reveals the positive feedback mechanism between online WOM and product sales. Our findings strongly support the value of c

25、onsidering the endogeneity of WOM and its interdependence with consumers’ consumption behavior. The notably different results obtained from 3SLS (the statistically more robust method) and OLS suggest that extant research

26、 using simple regression techniques may have drawn biased conclusions about the direction and magnitude o</p><p>　　Our findings also bring important extensions to previous research (Basuroy et al. 2003; Elia

27、shberg and Shugan 1997; Liu 2006) on the relationship among WOM volume ,WOM valence and box office sales. Previous research has been focusing on</p><p>　　the direct impact of WOM volume and valence on box of

28、fice revenues and find that most of the explanatory power comes from WOM volume not WOM valence (Liu 2006). Our study extends this approach by considering the interaction between WOM valence and WOM volume. We find that

29、while WOM valence does not directly affect revenue, higher WOM valence indirectly increases box office revenue by generating higher volume of WOM.</p><p>　　The contributions of this research to retail litera

30、ture on WOM are multifaceted. From the methodology perspective, we bring to light the importance of separating the effect of WOM as both a precursor and an outcome of sales. Our results also highlight</p><p>

31、;　　the importance of using a dynamic system and high-frequency data in studying the effect of WOM in the digital environment. From the managerial perspective, we show that WOM valence and WOM volume play different roles

32、in influencing product sales. We also show that time-series changes in WOM valence influences WOM volume which leads to higher product sales. Our findings support the idea that the online WOM process has a significant im

33、pact on sales, suggesting that businesses should</p><p>　　embrace and facilitate WOM activities.</p><p>　　There are a number of opportunities to extend the current research. One important and in

34、teresting extension of our research will be to investigate the consumer decision process under the influence of WOM information, especially in the digital</p><p>　　environment. In addition, not all WOM is eq

35、ual. Consumers need to distinguish the “true” and “honest” opinions from all kinds of feedback and recommendations on the web. Under such circumstances, how consumers choose their information source find trusted informat

36、ion sources will be of particular interest for future research.</p><p>　　Online user reviews are only one type of consumer-generated media. The recent explosive growth of popular online social communities (e

37、.g., www.YouTube.com, www.Flickr.com, and www.Digg.com) has generated a renewed interest in the Internet as a new medium for content generation and distribution. Different from online review sites we explored here, onlin

38、e social communities encourage interaction between users, which potentially changes the dynamics of WOM distribution. The modeling approach used i</p><p>　　Our analysis is, by necessity, restricted to online

39、 users who choose to post reviews and to post them on Yahoo! Movie. Thus, our estimates are conditioned on such a user population. While such a restriction does not necessarily bias the panel data estimation results, the

40、y should be interpreted as applying to a self-selected set of online users. In addition, this paper studies only the relationship between the postre lease WOM and sales. However, WOM certainly has existed before a movie’

41、s release</p><p>　　出處：Wenjing Duan,Bin Gu, Andrew B.The dynamics of online word-of-mouth and product sales—An empirical investigation of the movie industry[J]. Journal of Retailing ,2008(6): 233-242</p>

42、;<p><b>　　二、翻譯文章</b></p><p>　　標(biāo)題： 在線口碑和產(chǎn)品銷售動(dòng)態(tài)——對(duì)電影行業(yè)的實(shí)證調(diào)查</p><p><b>　　譯文：</b></p><p><b>　　介紹</b></p><p>　　口碑自人類社會(huì)開(kāi)始以來(lái)，已經(jīng)被公認(rèn)為

43、最有影響力的信息傳遞的資源(Godes and Mayzlin 2004; Maxham and Netemeyer 2002; Reynolds and Beatty 1999).然而，傳統(tǒng)的人際間口碑交流只有在有限的社會(huì)接觸范圍內(nèi)才會(huì)起作用，其影響隨時(shí)間和空間的推移迅速減弱(Bhatnagar and Ghose 2004; Ellison and Fudenberg 1995)。信息技術(shù)的進(jìn)步和在線社交網(wǎng)站的出現(xiàn)，深刻地改變了信息

44、傳播的方式，并已超過(guò)了傳統(tǒng)口碑的界限(Laroche et al. 2005)。另外，在一個(gè)或幾個(gè)朋友間傳播的短暫口碑，已經(jīng)被看作是能透過(guò)此看見(jiàn)世界的信息。因此，在線口碑在消費(fèi)者購(gòu)買決策中發(fā)揮著日益重要的作用。</p><p>　　在線口碑對(duì)零售商來(lái)說(shuō)，既是機(jī)遇也是挑戰(zhàn)。一方面，口碑為消費(fèi)者提供了可供選擇的信息源，從而降低了零售商通過(guò)傳統(tǒng)市場(chǎng)營(yíng)銷和廣告渠道影響消費(fèi)者的能力。先前的研究表明，口碑的各個(gè)方面會(huì)影響到零

45、售的銷量。一些研究發(fā)現(xiàn)，口碑量(Godes and Mayzlin 2004)和價(jià)數(shù)(Chevalier and Mayzlin 2006; Forman, Ghose, andWiesenfeld 2008)對(duì)產(chǎn)品銷量產(chǎn)生重大影響，而另一些研究發(fā)現(xiàn)，口碑量被看作是產(chǎn)品銷量的主要驅(qū)動(dòng)力(Chen, Wu, and Yoon 2004; Liu 2006).。另一方面，在線口碑為零售商提供一個(gè)接觸到消費(fèi)者和戰(zhàn)略影響消費(fèi)者意見(jiàn)的場(chǎng)所。近幾年

46、出現(xiàn)的傳聞證據(jù)表明，在線口碑已作為新的營(yíng)銷工具被成功應(yīng)用(Dellarocas2003).。不同于傳統(tǒng)的市場(chǎng)營(yíng)銷效果，口碑效應(yīng)的獨(dú)特之處在于口碑和產(chǎn)品銷量的積極反饋進(jìn)程，也就是說(shuō)，口碑會(huì)提高產(chǎn)品銷量，反過(guò)來(lái)產(chǎn)生更多的口碑和更多的產(chǎn)品銷量。積極的反饋機(jī)制表明，口碑不僅是消費(fèi)者購(gòu)買的驅(qū)動(dòng)力，而且是零售銷量的結(jié)果( Godes and Mayzlin 2004; Srinivas</p><p>　　我們選擇了電影產(chǎn)業(yè)

47、作為研究范疇。因?yàn)闃I(yè)界專家一致認(rèn)為，口碑是能令電影持續(xù)發(fā)揮其功效的一個(gè)關(guān)鍵因素，從而實(shí)現(xiàn)最終財(cái)務(wù)上的成功(Elberse and Eliashberg 2003)。此外，電影界迄今為止在文學(xué)營(yíng)銷口碑上受到重視，這使得我們的成果能和之前的研究進(jìn)行深入的比較。然而，我們注意到電影作為一種特殊的體驗(yàn)產(chǎn)品類型，其產(chǎn)業(yè)的結(jié)果并不一定能夠推廣到其他零售行業(yè)。相反，我們的目標(biāo)是利用電影產(chǎn)業(yè)作為一個(gè)內(nèi)容來(lái)強(qiáng)調(diào)考慮零售銷量和在線口碑的動(dòng)態(tài)和他們之間關(guān)系的

48、重要性，并說(shuō)明設(shè)置的聯(lián)立方程方法的正確性。我們發(fā)現(xiàn)，一部電影的票房收入和口碑效價(jià)明顯地影響著口碑量，口碑量反過(guò)來(lái)會(huì)帶來(lái)更高的票房和更好的成績(jī)。我們的研究結(jié)果澄清了早期研究中用戶評(píng)分對(duì)票房收入影響的博弈。我們表明，用戶評(píng)分不會(huì)直接影響票房收入。但是，他們會(huì)通過(guò)口碑間接影響到票房收入。</p><p>　　此外，我們的研究還證實(shí)了在線口碑不僅是先導(dǎo)，而且是產(chǎn)品銷量的結(jié)果。我們表明， 忽視口碑的雙重性質(zhì)會(huì)導(dǎo)致錯(cuò)誤的結(jié)果

49、。</p><p>　　電影產(chǎn)業(yè)的在線口碑有多種表現(xiàn)形式，包括在線評(píng)論，討論版，聊天室，博客，維基和其他。在此研究中，我們側(cè)重于在線用戶評(píng)論，因?yàn)閿?shù)據(jù)顯示，在電影產(chǎn)業(yè)中，在線用戶評(píng)論比其他口碑交流更加普遍。除了量之外，在線用戶評(píng)論和其他形式的口碑之間的重要微妙區(qū)別在于，用戶評(píng)論通常反映了用戶的經(jīng)歷和滿意度，這被看作是產(chǎn)品的信息源之一(Chen and Xie 2004; Li and Hitt 2008).。與此

50、同時(shí)，口碑的其他類型，比如較能反映消費(fèi)者的預(yù)期的在線社區(qū)網(wǎng)站的討論，深受社會(huì)結(jié)構(gòu)的影響(Gopal et al. 2006; Liu 2006)。</p><p>　　本文的其余部分組織如下。下一節(jié)將提供支撐我們概念框架和研究模型的文獻(xiàn)。然后，我們描述我們的數(shù)據(jù)來(lái)源和經(jīng)驗(yàn)?zāi)Ｐ图捌漕A(yù)測(cè)。之后，將描述和討論主要的調(diào)查結(jié)果，以及討論的意義、限制及未來(lái)的研究。</p><p><b>　

51、　實(shí)證模型詳述</b></p><p>　　我們的實(shí)證模型發(fā)展遵循以下考慮。首先，由于我們對(duì)電影票房收入和口碑驅(qū)動(dòng)力感興趣，我們構(gòu)建了兩個(gè)相互依存的系統(tǒng)，一個(gè)是每日收入作為因變量方程（收入方程），另一個(gè)是口碑作為因變量的方程（口碑方程）。我們假設(shè)在每個(gè)階段（例如，一天），兩個(gè)方程的錯(cuò)誤是相關(guān)的，這意味著不包括在我們模型的因素會(huì)同時(shí)影響著電影的收入和口碑。</p><p>　　其

52、次，認(rèn)識(shí)到消費(fèi)者對(duì)電影的選擇行為和口碑之間相互作用會(huì)超越同期(Elberse and Eliashberg 2003)，我們發(fā)展了一個(gè)動(dòng)態(tài)方程系統(tǒng)。在收入方程中，不僅包括同期的每日口碑量，而且還有滯后的因素。同樣，在口碑方程中，滯后的收入因素也同樣存在著。這樣的規(guī)范有助于識(shí)別兩個(gè)聯(lián)立方程系統(tǒng)，因?yàn)闇笠蛩貙儆诿總€(gè)方程的外在變量。此外，繼現(xiàn)存的研究，我們?cè)谀Ｐ椭惺褂镁€性方程(e.g., Elberse and Eliashberg 200

53、3; Liu2006)。線性方程符合消費(fèi)者決策過(guò)程的各個(gè)階段，一部電影的銷量可以看作是應(yīng)用到消費(fèi)群的一系列條件概率。一個(gè)數(shù)可轉(zhuǎn)換成經(jīng)驗(yàn)估計(jì)的線性關(guān)系模型，此外，線性方程中回歸變量的平滑分布，以及線性形式的估計(jì)系數(shù)可直接反應(yīng)非獨(dú)立和獨(dú)立變量的彈性。第三，為了控制更多可影響影響電影收入和口碑的特有因素，如預(yù)算、市場(chǎng)營(yíng)銷、明星和其他(Basuroy et al. 2003; Elberse and Eliashberg 2003; Liu 2

54、006)，我們把在模型中通過(guò)增加特定虛擬變量所帶來(lái)的固定效果包括在內(nèi)。固定效果捕獲了任何變化因素，包括內(nèi)在的電影特色，</p><p>　　意義、限制和未來(lái)研究</p><p>　　我們的模型詳細(xì)說(shuō)明了雙重因果關(guān)系，并揭示了在線口碑和產(chǎn)品銷量之間的積極反饋機(jī)制。我們的研究有力地支持了考慮口碑的內(nèi)在性質(zhì)和消費(fèi)者消費(fèi)行為之間的相互依存關(guān)系。從3SLS（更強(qiáng)大的統(tǒng)計(jì)方法）和OLS的明顯不同得出，

55、利用簡(jiǎn)單回歸技術(shù)的現(xiàn)存研究可能會(huì)得出關(guān)于口碑效應(yīng)方向和大小的偏見(jiàn)結(jié)論。我們的結(jié)果驗(yàn)證了我們關(guān)于在線用戶評(píng)論數(shù)量和零售銷量雙贏關(guān)系的斷言。</p><p>　　我們的研究也是對(duì)先前關(guān)于口碑量、口碑效價(jià)、票房銷量之間關(guān)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響和票房收入價(jià)的效價(jià)，并發(fā)現(xiàn)主要的

56、解釋力量來(lái)自于口碑量而不是口碑效價(jià)(Liu 2006)。我們的研究拓展了考慮口碑量和口碑效價(jià)相互作用的方法，我發(fā)現(xiàn)，雖然口碑效價(jià)不能直接影響收入，但更高的口碑效價(jià)能通過(guò)產(chǎn)生更多的口碑量間接提高票房收入。此研究對(duì)口碑文獻(xiàn)的貢獻(xiàn)是多方面，從方法論的角度來(lái)看，我們揭示了分離先導(dǎo)和銷售結(jié)果的口碑效應(yīng)的重要性。同時(shí)，我們的結(jié)果強(qiáng)調(diào)了，在數(shù)字環(huán)境中，使用動(dòng)態(tài)系統(tǒng)和高頻數(shù)據(jù)對(duì)研究口碑效應(yīng)的重要性。從管理角度來(lái)看，我們說(shuō)明了口碑效價(jià)和口碑量在影響產(chǎn)品銷

57、量中扮演著不同的角色。我們還說(shuō)明，口碑效價(jià)時(shí)間序列的變化影響著能帶來(lái)更高銷量的口碑量。我們的研究支持在線口碑進(jìn)程對(duì)銷量有重要影響的想法，這表明企業(yè)應(yīng)支持和豐富口碑活動(dòng)。</p><p>　　有很多機(jī)會(huì)去延伸現(xiàn)有的研究，我們研究中的一項(xiàng)重要和有趣的擴(kuò)展是去調(diào)查消費(fèi)者在口碑信息，尤其是數(shù)字環(huán)境中的決策過(guò)程。此外，并非所有的口碑都是平等的，消費(fèi)者需要從網(wǎng)上所有反饋和建議中尋找到真實(shí)和誠(chéng)實(shí)的意見(jiàn)。在這種情況下，消費(fèi)者如何

58、從信息源中找到可信賴的信息，是今后研究中特別有趣的。</p><p>　　在線用戶評(píng)論只是一種，連接消費(fèi)者的媒介。作為一項(xiàng)新媒介的生成和分布的在線社區(qū)(e.g., www.YouTube.com, www.Flickr.com, and www.Digg.com)受歡迎程度的急劇增長(zhǎng)，在互聯(lián)網(wǎng)上引起了新的關(guān)注。與在線網(wǎng)站評(píng)論不同，我們?cè)谶@里探討，在線社交社區(qū)鼓勵(lì)用戶之間的互動(dòng)，這可能會(huì)改變口碑的動(dòng)態(tài)分布。因此，本

59、次研究中使用的模型方法可能不能充分地反映所研究的內(nèi)容。對(duì)新媒介中在線口碑效應(yīng)的描述和確定的新研究，有利于我們理解在線連接消費(fèi)者的媒介對(duì)市場(chǎng)營(yíng)銷和零售戰(zhàn)略所產(chǎn)生的影響。</p><p>　　我們的分析，根據(jù)需要對(duì)選擇發(fā)帖評(píng)論和在雅虎評(píng)論電影的在線用戶進(jìn)行限制。因此，我們的研究是對(duì)這樣一個(gè)有條件的用戶群的估計(jì)。雖然這樣的限制不一定有偏于面板數(shù)據(jù)的預(yù)測(cè)結(jié)果，他們應(yīng)該被解釋為運(yùn)用到一個(gè)在線用戶的自我選擇設(shè)置。此外，本為只

60、是對(duì)張貼發(fā)布口碑和銷量之間關(guān)系的研究。不過(guò)，口碑在電影發(fā)布前是肯定存在的，電影制片廠作出各種營(yíng)銷努力以促進(jìn)口碑(Liu 2006)。關(guān)于雅虎電影的一項(xiàng)詳細(xì)調(diào)查顯示了預(yù)發(fā)布口碑和發(fā)布口碑之間重要和有趣的區(qū)別。我們注意到，在雅虎電影討論版的預(yù)發(fā)布口碑活動(dòng)中心，帖子主要反映了消費(fèi)者期望。同時(shí)，在雅虎電影用戶評(píng)論網(wǎng)站的發(fā)口碑貼中心，帖子主要反映消費(fèi)者的滿意度和產(chǎn)品體驗(yàn)。這種差別表明了在預(yù)發(fā)布口碑和發(fā)布口碑之間，確實(shí)存在著不同的機(jī)制。因此，對(duì)現(xiàn)有