|
on Marketing |
|
Issue of 2024‒02‒26
two papers chosen by |
| By: | Shawn Berry |
| Abstract: | The use of online reviews to aid with purchase decisions is popular among consumers as it is a simple heuristic tool based on the reported experiences of other consumers. However, not all online reviews are written by real consumers or reflect actual experiences, and present implications for consumers and businesses. This study examines the effects of fake online reviews written by artificial intelligence (AI) on consumer decision making. Respondents were surveyed about their attitudes and habits concerning online reviews using an online questionnaire (n=351), and participated in a restaurant choice experiment using varying proportions of fake and real reviews. While the findings confirm prior studies, new insights are gained about the confusion for consumers and consequences for businesses when reviews written by AI are believed rather than real reviews. The study presents a fake review detection model using logistic regression modeling to score and flag reviews as a solution. |
| Date: | 2024–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2401.11345&r=mkt |
| By: | Yigit Oezcelik; Michel Tolksdorf |
| Abstract: | Inaccurate online ratings can be harmful to both consumers and firms. We perform an experiment to assess the effect of the anchoring bias on consumer ratings. Our rating task is a framed variation of the slider task. We diverge from the literature by implementing non-numerical (visual) anchors. We compare three anchoring conditions, with either high, low, or socially derived anchors present, against two control conditions – one without anchors and an unframed slider task. High and socially derived anchors lead to significant overrating compared to both control conditions. We find no difference between low anchors and the control condition without anchors, whereas both exhibit overrating compared with the unframed slider task. Participants place higher trust in socially derived anchors compared with high and low anchors. When there is a social context, the trust participants exhibit towards the socially derived anchors explains the anchoring bias. |
| Keywords: | Anchoring, online ratings, laboratory experiment |
| JEL: | C91 D80 D91 |
| Date: | 2023–12 |
| URL: | http://d.repec.org/n?u=RePEc:liv:livedp:202319&r=mkt |