| Article ID: | iaor20112097 |
| Volume: | 50 |
| Issue: | 3 |
| Start Page Number: | 627 |
| End Page Number: | 635 |
| Publication Date: | Feb 2011 |
| Journal: | Decision Support Systems |
| Authors: | Bose Indranil, Liu Ling, Hu Nan, Gao Yunjun |
| Keywords: | decision support, bookselling, fraud |
Built upon the discretionary accrual‐based earnings management framework, our paper develops a discretionary manipulation proxy to study the management of online reviews. We reveal that fraudulent review manipulation is a serious problem for 1) non‐bestseller books; 2) books whose reviews are classified as not very helpful; 3) books that experience greater variability in the helpfulness of their online reviews; and 4) popular books as well as high‐priced books. We also show that review management decreases with the passage of time. Just like fraudulent earnings management, manipulated online reviews reflect inauthentic information from which consumers might derive wrong valuation especially for books with the above characteristics and be persuaded to purchase the wrong item. The findings from this research sound a note of caution for all consumers that make use of online reviews of books for making purchases and encourage them to delve deeper into the reviews without getting trapped in their fraudulent manipulation.