Text Performance on the Vine Stage? The Effect of Incentive on Product Review Text Quality

• 成果类型: Review
• 署名作者: Qiao, Dandan; Rui, Huaxia
• 署名单位: National University of Singapore; University of Rochester
• 刊物名称: INFORMATION SYSTEMS RESEARCH
• ISSN/ISSBN: 1047-7047
• DOI: 10.1287/isre.2022.1146
• 发表日期: 2023
• 页码: 676-697
• 关键词: local coherence

摘要

Incentivized reviews have become increasingly prevalent on product review sites such as Amazon. Whereas outright fake reviews are clearly unacceptable and should be removed from review platforms, reviews by incentivized consumers with otherwise authentic product experiences fall into a gray area. On the one hand, many critics and researchers have warned of their harm by pointing out their biased ratings. On the other hand, these reviewsmight complement organic reviews with review text of higher quality. The current paper studies whether incentivized reviews on Amazon are more coherent and offer richer detail. We use Amazon's platform-incentivized reviews, known as Vine reviews, for our primary sample and use seller-incentivized reviews for checking robustness. Estimations from a two-way fixed-effect model consistently show that incentivized reviews do compensate for their reduced impartiality through better text quality, measured by discourse coherence and level of relevant detail. This finding is further supported by a randomized experiment using Amazon Mechanical Turk. Hence, current literature findings on the poor quality of text of incentivized reviews, based on review length and lexical complexity, portray only an incomplete picture of incentivized reviews. Given that numerical ratings for products via incentivized reviews are likely biased whereas their text content is of high quality, a natural way to embrace incentivized reviews is to keep their text content, suppress their numerical ratings, and always highlight the label of incentivized reviews.