The need to trust: How features of the forecasted weather influence forecast trust

Abstract

Although it is now clear that trust affects risk perception, the opposite may be true as well—risk perception may affect trust. This research investigated the effect of several previously unexamined risk-related predictors of forecast trust—forecast severity, forecast inconsistency, and the perceiver's familiarity with the forecasted weather. Each of these factors was manipulated experimentally using hypothetical forecasts and scenarios. On each trial, participants saw two forecasts—one early and one late in the week. Half of trials concerned thunderstorms, the other half snow. Half of participants lived in a region where each of these phenomena occurred more than the other. Forecasts differed in level of severity and whether both forecasts in the sequence predicted the same level of severity (consistent) or not (inconsistent). Multilevel analyses revealed that characteristics of the forecast including increased severity, consistency, and decreased familiarity lead to greater trust in the forecast. Additionally, perceptual variables such as an increased estimated likelihood of severe weather, expectation of danger, and harm were also associated greater trust. Moreover, analyses using generalized estimating equations (GEE) revealed that greater forecast severity, inconsistency and greater user trust lead to a higher likelihood of taking risk mitigation action. This work contributes to our understanding of the conditions that give rise to trust in weather forecasts and how that translates into protective action.