A long time ago, in a lab far, far away ...
An experiment in uncertainty
This interesting study tested the hypothesis that giving people a handle on the uncertainty in numerical measures does not affect their trust in the source. The experiment was conducted using a BBC News article on unemployment. The headline finding was that uncertainty communication does not reduce trust. However, I draw attention to this sentence An ANOVA showed a significant main effect of uncertainty communication on perceived uncertainty [ F (2, 1526) = 4.67, P = 0.01; η 2 = 0.006 ]. So the effect of uncertainty communication on perceived uncertainty was absolutely minuscule (partial-eta 0.006)! The findings on trustworthiness then need to be taken with a large pinch of salt. I would then revise the take-home message of this paper to be - the methods used to communicate uncertainty in this study were pretty ineffective.
Comments
Post a Comment