
The Effect of Online Reviews Credibility Perception on Product Sales
Qiang WANG, Wen ZHANG, Jian LI, Zhenzhong MA
China Journal of Econometrics ›› 2023, Vol. 3 ›› Issue (2) : 513-530.
The Effect of Online Reviews Credibility Perception on Product Sales
With the rapid development of E-commerce, online reviews are becoming an important resource to support consumer purchase decision making. This makes the online fraudulent reviews pervasive in the E-commerce platform resulting in great loss for both online consumers and merchants. Based on the persuasion knowledge theory, this study explores the effect of online reviews credibility perception on product sales. On the one hand, we explore the effect of online reviews credibility perception on consumer purchase decisions with adoption of the system GMM model. On the one hand, we extract review text features from online reviews and to study its effect on consumers' credibility perception of online reviews with adoption of the fractional logit model. The result shows that online reviews credibility perception moderates the effect of consumers' information processing of online reviews on product sales. The online review features as referring, contextual, detailing, and argument structuring have significant effect on online reviews credibility perception. This study has great managerial implications for both platforms and vendors to improve their practice on online review management.
fraudulent online reviews / text analytics / review credibility perception / fractional logit / system GMM {{custom_keyword}} /
表1 影响评论感知真实性的评论文本表现形式 |
评论文本表现形式 | 释义 |
参考性 | 评论中涉及评论者和被评论对象信息的描述 |
语境嵌入 | 评论中对情景和语境相关时间和空间线索的描述 |
细节 | 评论中对产品或服务细节的描述 |
论点结构化 | 评论内容的一致性和连贯性 |
表2 收集的产品及评论相关数据 |
相关数据类型 | 描述 | 变量符号 |
销量 | 产品的销量排名 | Log(sales_rank) |
评论效价 | 产品的评论评分 | Rating |
评论情感 | 根据评论文本内容计算的评论情感强度 | Sentiment |
评论感知真实性 | 根据评论文本内容计算的评论感知真实性 | Credibility |
价格 | 产品的价格 | Price |
评论量 | 产品的评论总量 | Review volume |
被认证购买比率 | 产品所有评论中被认证购买的评论占比 | Verified |
物流评分 | 产品的物流评分 | Logistics |
产品上架周数 | 产品自发布以来的累计周数 | Weeks |
表3 消费者对在线商品评论感知真实性的标注结果 |
类别 | 均值 | 标准差 | 最小值 | 最大值 |
标注者 | 0.66 | 0.08 | 0.06 | 0.89 |
评论 | 0.69 | 0.05 | 0.15 | 0.83 |
表4 评论文本表现形式的测量 |
变量 | 测量 | 变量符号 |
参考性 | 评论中人称代词的比率 | Referencing |
语境嵌入 | 评论中时间及空间类型词的比率 | Contextual |
细节 | 评论中细节词的比率 | Detailing |
论点结构化 | 评论中感知过程词的比率 | Argument |
表5 消费者对在线商品评论感知真实性的标注结果 |
变量 | 变量符号 | 均值 | 标准差 | 最小值 | 最大值 |
销量 | Log(sales_rank) | 3.73 | 0.76 | 1.79 | 5.22 |
评论效价 | Rating | 4.80 | 0.03 | 4.7 | 4.9 |
评论情感 | Sentiment | 3.30 | 0.29 | 2.7 | 3.76 |
评论感知真实性 | Credibility | 0.68 | 0.07 | 0.12 | 0.85 |
价格 | Price | 226.27 | 78.18 | 198 | 243 |
评论量 | Review volume | 62.08 | 69.46 | 9 | 327 |
被认证购买比率 | Verified | 80.34 | 6.37 | 74.28 | 97.48 |
物流评分 | Logistics | 4.50 | 0.01 | 3 | 6 |
产品上架周数 | Weeks | 207.16 | 27.58 | 8 | 388 |
表6 评论感知真实性对产品销量影响的模型回归结果 |
变量 | 模型1 | 模型2 | 模型3 |
Rating | 0.240*** | 0.144*** | |
Sentiment | 0.888*** | 0.400*** | |
Sentiment | |||
Credibility | 0.305*** | 0.506*** | |
Credibility | 0.001** | ||
Credibility | |||
Credibility | 0.096*** | ||
Price | |||
Review volume | 0.412*** | 0.046*** | 0.597*** |
Verified | 0.394*** | 0.205*** | 0.203*** |
Logistics | 0.428*** | 0.190*** | 0.132*** |
Weeks | 0.289** | 0.277*** | 0.127*** |
注: ***p < 0.001, **p < 0.01, *p < 0.05. |
表7 评论感知真实性对产品销量影响的模型回归结果 |
变量 | 模型4 | 模型5 |
Referencing | 0.405*** | |
Contextual | 0.471*** | |
Detailing | 0.518*** | |
Argument | 0.723*** | |
Length | 0.256** | 0.227** |
Helpfulness | 0.545*** | 0.134** |
Verified | 0.237** | 0.147** |
注: ***p < 0.001, **p < 0.01, *p < 0.05. |
表8 评论感知真实性预测模型在不同均方误差下的回归模型稳健性检验结果 |
变量 | MSE=0.46 | MSE=0.48 | MSE=0.50 |
Rating | 0.144*** | 0.181** | 0.218*** |
Sentiment | 0.400*** | 0.113** | 0.170*** |
Sentiment | |||
Credibility | 0.506*** | 0.682*** | 0.890*** |
Credibility | 0.001** | 0.001** | 0.001** |
Credibility | |||
Credibility | 0.096*** | 0.431*** | 0.235*** |
Price | |||
Review volume | 0.597*** | 0.960*** | 0.834*** |
Verified | 0.203*** | 0.151*** | 0.167*** |
Logistics | 0.132*** | 0.231*** | 0.151*** |
Weeks | 0.127*** | 0.278*** | 0.240*** |
注: ***p < 0.001, **p < 0.01, *p < 0.05. |
洪永淼, 汪寿阳, 数学、模型与经济思想[J]. 管理世界, 2020, 36 (10): 15- 27.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
洪永淼, 汪寿阳, 大数据、机器学习与统计学: 挑战与机遇[J]. 计量经济学报, 2021, 1 (1): 17- 35.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
洪永淼, 汪寿阳, 大数据如何改变经济学研究范式?[J]. 管理世界, 2021, 37 (10): 40- 55.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
李建平, 王军, 冯倩倩, 孙晓蕾, 基于多元驱动因素的主权CDS利差预测研究[J]. 计量经济学报, 2021, 1 (2): 62- 76.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
李建平, 吴登生, 郝俊, 技术驱动的管理科学与工程研究[J]. 中国管理科学, 2022, 30 (5): 9- 14.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
余乐安, 基于人工智能的预测与决策优化理论和方法研究[J]. 管理科学, 2022, 35 (1): 60- 66.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
余乐安, 范常容, 马月明, 基于混合多分形小波的国际油价多期预测研究[J]. 计量经济学报, 2021, 1 (3): 612- 623.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
张文, 王强, 李健, 马振中, 谢锐, 在线商品虚假评论发布动机及形成机理研究[J]. 中国管理科学, 2022, 30 (7): 176- 188.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
Chaiken S, Liberman A, Eagly A H H. 1989. Heuristic and Systematic Information Processing Within and Beyond the Persuasion Context[M]//Uleman J S, Bargh J A. Unintentent Thought. NewYork: Guilford Press: 212-252.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
Trustpilot. 2020. The Critical Role of Reviews in Internet Trust[R]. France: Trustpilot Platform.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |