Are Online Reviews Just Noise? The Truth, the Whole Truth, or Only the Partial Truth?

• Main Authors: HU, Nan, LIU, Ling, Sambamurthy, Vallabh, CHEN, Bin
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2009
• Subjects: Computer Sciences; E-Commerce; Numerical Analysis and Scientific Computing
• Online Access: https://ink.library.smu.edu.sg/sis_research/338
• Institution: Singapore Management University

Given consumers are increasingly depending on information from non-traditional information channel, such as online reviews, to make purchase and valuation decision. It is crucial and timely to understand the quality aspect of information released through this new channel. Based on data collected from Amazon, Barnes & Noble, and Downloads.com, we analytically and empirically investigate online review manipulation behavior and conclude that 1) Vendors, publishers, and writers consistently manipulate online consumer reviews. For a firm already joining in the manipulation club, its manipulation strategy is a monotonically decreasing function of that firm's true quality and the average consumer rating of the products that firm is selling. Thus, manipulation decreases the informativeness of online reviews. 2) Consumers understand the existence of manipulation, but can only partially correct such manipulation bias based on their expectation of the overall level of manipulation. Hence, vendors are able to change the outcomes of the results by manipulating online consumer reviewers. 3) Not all vendors engage in the practice of online consumer opinion manipulation. Over the long run, the persistence of this practice might result in a market failure because A) Online reviews are less informative; and B) Consumers punish all vendors, regardless of whether they are involved in manipulation. Hence, vendors that conduct manipulation are better off. 4) We document that at the early stage after an item is released to the Amazon market, both price and reviews serve as quality indicators. Thus, at this stage a higher price leads to an increase in sales instead of a decrease in sales. Finally, 5) we show that on average there is higher level of manipulation on Barnes & Noble than on Amazon. Thus, like earning management, online reviews might reflect cooked information based on which consumers might derive the wrong valuation. We call for collective thinking within this community, including technical vendors and business players, to build a better online system to fight against this practice.