Understanding Online Trust and Information Behavior Using Demographics and Human Values

Abstract

In the aftermath of the 2016 U.S. Presidential Election, the role of social media in influencing the dissemination of information is a timely and critical issue. To understand how social media-based information, misinformation, and disinformation work in practice, it is critical to identify factors that can predict social media users’ online information behavior. To this end, we designed an experiment to study the influence of the independent variables, demographics, and human values, on the dependent variables, social media users’ observed trust behavior, self-reported trust behavior, and information behavior. We report the statistically significant results of these comparisons; for example, we found that liberals were more likely to trust mainstream media (p < 0.05) and scientific journals (p < 0.05) and to state that the content of the linked pages influenced their trust (p < 0.01) than moderates; for values, we found that participants who more highly valued security were more likely to trust mainstream media articles (p < 0.05), to notice the presence or absence of hyperlinks, and to click on fake news articles (p < 0.05). Ultimately, both demographics and values can be used to predict online trust and information behavior; while demographics are commonly captured or predicted in online marketing, values represent a much less tapped opportunity to predict social media users’ online trust and information behavior.