The impact of uncertainty communication on trust in its sources: a systematic review and meta-analysis

Abstract Uncertainty is inherent in evidence, yet debates persist about whether communicating this uncertainty affects audiences’ trust. We synthesized experimental research on the effects of uncertainty communication—relative to no uncertainty communication—on trust in the source. A random-effects meta-analysis of 28 studies revealed no overall effect, with the vast majority of studies yielding null results (k = 23). The result held when restricting the analysis to studies without major methodological limitations, that is, studies that did not introduce uncertainty in the control condition and only manipulated the intervention with uncertainty information (k = 7). Uncertainty communication neither systematically undermined nor enhanced trust. Few moderating factors have been identified by the included studies, and an exploratory subgroup analysis found no moderating effects. Furthermore, a systematic analysis of stimulus materials uncovered conceptual and methodological issues in the field. For example, control conditions varied in whether they represented certainty or merely the absence of uncertainty. Moreover, most studies relied on fictitious communication contexts with no real-world communicators. We discuss the findings’ implications and conclude by offering guiding questions for future research to enhance methodological and conceptual rigour in the field.