Two studies on individuals' perception on online disinformation

Abstract

Online disinformation has become a relatively common phenomenon on several platforms including social media platforms, ecommerce platforms and news platforms, and causes several serious consequences. In this paper, I explore two of its primary damaging manifestations, namely fake reviews of goods or services, which can mislead customers and potentially diminish firms' long-term profits, and fake news, which are intentionally and verifiably false to mislead readers. In the first study, I focus on the effect of negative rating deviation on perceived review manipulation and explore the boundary conditions. Existing studies mainly focus on the detection and impact of fake reviews but do not investigate customers' perception of review legitimacy. By introducing a new concept — perceived review manipulation — which evaluates the extent to which a customer regards the opinions or recommendations of a review to be misleading or manipulated, this study investigates the effect of reviews that deviate from average ratings on perceived review manipulation and explores the moderating effect of review content concreteness and reviewer rating distribution. This study also examines whether perceived review manipulation functions as a mechanism that mediates the effect of deviant ratings on perceived review helpfulness/adoption. By conducting two online randomized experiments and one field study on Yelp, the findings suggest: (1) reviews with deviant ratings are more likely to be perceived (filtered) as review manipulation; (2) customers (platforms) are more likely to perceive (classify) reviews with deviant ratings as review manipulations when review content is abstract rather than concrete; (3) customers are more (less) likely to perceive reviews with deviant ratings as review manipulation when reviewer rating distribution is negative (positive); (4) platforms are more (less) likely to filter reviews with deviant ratings as review manipulation when the skewness of reviewer rating distribution is large (small); (5) perceived review manipulation mediates the effect of reviews with deviant ratings on perceived review helpfulness/adoption. In the second study, I intend to investigate individuals' ability to distinguish fake news from real news in normal situations and in the situations under which different arousal level and personal involvement exist. Fake news has penetrated to individuals' life especially when a particular epidemic event occurs. The reason is that the occurrence of some epidemic events is more likely to motivate individuals to evaluate news based on their sentiments rather than rationality, thus likely lowering individuals' ability to distinguish between real news and fake news. This study focuses on the role of two dimensions of sentiments — arousal level and personal involvement — on individuals' judgement on real news and fake news. By conducting two online randomized experiments in the context of COVID-19, this study finds that individuals have the ability to distinguish fake news and real news in normal situations, while they are more likely to trust or share fake news when news content can trigger high arousal level. For individuals with a high involvement toward COVID-19 news, their trusting perception toward fake news whose content triggering high arousal level becomes even higher. The findings have implications for academics and practitioners. Theoretically, the findings contribute to online review literature by proposing the new concept of perceived review manipulation and identifying the factors that can influence perceived review manipulation as well as fake news literature by introducing arousal level and personal involvement to the new context and better understanding individuals' evaluations on fake news. Practically, the findings provide new insights for ecommerce platforms and news platforms about the emphasis on online disinformation regulation.