Trusting expert- versus user-generated ratings online: The role of information volume, valence, and consumer characteristics

Keywords: User-generated content Online ratings Social media Online social influence Trust Credibility a b s t r a c t The warranting principle, signaling theory, and theories of informational social influence suggest conditions when either user-generated information, or information originating from traditional experts, might be privileged online. A random sample of 1207 U.S.-based adults with Internet access completed an experiment that manipulated the source, volume, and valence of online movie ratings in order to test predictions derived from these perspectives. Results indicated that ratings volume is positively associated with trust of, reliance on, and confidence in user-generated content, as well as the congruence between one's own and others' opinions; that ratings source and volume interact to impact credibility perceptions, reliance on user-generated information, and opinion congruence, such that people tend to favor experts when there is low information volume, but favor user-generated information under conditions of high information volume; and that people's opinions and behavioral intentions converge with the online ratings information to which they are exposed. In addition, these effects apply more strongly to people more conversant with user-generated content. Results indicate important theoretical extensions by demonstrating that social information online may be filtered through signals indicating its veracity, which may not apply equally to all social media users. Among the most important consequences of digital media in recent decades is that they provide an environment where individuals are routinely exposed to a tremendous range of opinions, across a broad diversity of sources. Contributing to this informational fecundity is a dramatic rise in ''user-generated content,'' where individuals not only consume, but also produce many of the information resources available online (Bruns, 2008; Ochoa & Duval, 2008). The user-generation of information is a potentially transforma-tive development that nonetheless co-exists with more traditional and well-established information sources that have also been changed in important ways by the capacity of networked technologies. The relative credibility of these information forms, or the degree to which each can be believed (Hovland, Janis, & Kelley, 1953), remains unclear. Given increasing reliance on online information generally, and user-generated content specifically, accurate credibility evaluations are critical since they impact people's attitudes and behaviors, and because information consumers today often make assessments without the aid of traditional credibility cues (Metzger & Flanagin, 2008). To explore these issues, we examine various forms of information provision from multiple sources by considering the theoretical mechanisms suggesting when and how each should …