Communicating Data Uncertainty: Multi-Wave Experimental Evidence for UK GDP
60 Pages Posted: 27 Dec 2021 Last revised: 14 Jul 2022
Date Written: July 13, 2022
Abstract
Economic statistics are commonly published without estimates of their uncertainty. We conduct two waves of a randomized controlled online experiment to assess if and how the UK public understands data uncertainty. A control group observes only the point estimate of GDP. Treatment groups are presented with alternative qualitative and quantitative communications of GDP data uncertainty. We find that most of the public understands that GDP numbers are uncertain. Quantitative communications of data uncertainty help align the public’s subjective probabilistic expectations of data uncertainty with objective estimates, but do not decrease trust in the statistical office.
Keywords: Experiments, Data Uncertainty, Uncertainty Communication, Data Revisions
JEL Classification: C82, E01, D80
Suggested Citation: Suggested Citation