›› 2018, Vol. 30 ›› Issue (4): 94-105.

• 市场营销 • 上一篇    下一篇

基于内生性考虑的网络口碑离散度变异系数特征研究

谢光明, 邱冬冬, 蒋玉石   

  1. 西南交通大学经济管理学院, 成都 610031
  • 收稿日期:2016-03-24 出版日期:2018-04-28 发布日期:2018-04-22
  • 通讯作者: 蒋玉石(通讯作者),西南交通大学经济管理学院副教授,博士生导师,博士。
  • 作者简介:谢光明,西南交通大学经济管理学院博士研究生;邱冬冬,西南交通大学经济管理学院硕士研究生
  • 基金资助:

    国家自然科学基金项目(71572156);四川省教育厅农业特色品牌开发与传播研究中心项目(CAB1501;CAB1505)。

A Research on the e-WOM's Dispersive Coefficient of Variation from the Perspective of Endogeneity

Xie Guangming, Qiu Dongdong, Jiang Yushi   

  1. School of Economics and Management, Southwest Jiaotong University, Chengdu 610031
  • Received:2016-03-24 Online:2018-04-28 Published:2018-04-22

摘要:

近年来,网络口碑研究日益推进,其中以网络口碑离散度尤甚。然而,已有文献多从聚合层面围绕网络口碑离散度与销量间的关系展开研究,缺乏对网络口碑离散度明确的概念界定,忽略了内生性问题在离散度特征上的讨论。本文围绕“网络口碑离散度”这一核心概念,引入变异系数对样本间离散程度差异进行讨论,并基于同一品类下产品属性及发布时间视角构建特征分析模型。之后,通过亚马逊(中国)网真实的消费者星级数据进行模型的实证分析。结果发现,网络口碑星级分布以一种常态化的“J”型非正态分布存在,即使是同一产品类别口碑离散度变异系数也存在明显差异,内生性因素对结果有着显著影响,其中享乐属性产品较之实用属性产品拥有较高的口碑离散度变异系数,产品发布时间与口碑离散度变异系数呈正向变化关系。实证结果丰富了网络口碑离散度的理论研究,也为后续口碑离散度影响研究提供了新的参考视角。

关键词: 离散度变异系数, 产品属性, 产品发布时间, 内生性, 图书

Abstract:

In recent years, researches on electronic word-of-mouth (e-WOM), especially the e-WOM dispersion, are becoming increasingly deeper. However, most of the existing researches explore the relationship between the dispersion and sales from the perspective of polymerization, with e-WOM dispersion not clearly defined and endogenous issues of dispersion not taken into consideration. With e-WOM dispersion as the core concept, we introduce coefficient of variation to discuss the disparity in dispersion degree across samples, and construct a model to analyze dispersion characteristic of the same product category based on product attributes and release time. Using the rating data of Amazon China, we find that the coefficient of variation of dispersion is significantly different even among the same product category, presenting a J-shaped abnormal distribution. This means that endogenous factors have significant influence. We also reveal that hedonic attribute has a higher coefficient of variation of e-WOM dispersion than utilitarian attribute, and that the relationship between product release time is positively related to coefficient of variation of e-WOM dispersion. The study enriches the existing researches on e-WOM dispersion and contributes to the understanding of the magnitudes of e-WOM effects.

Key words: coefficient of variation of e-WOM dispersion, product attribute, product release time, endogeneity, books