Belief Displacement in European GM-Food Safety Discourse (CWIAL Report No. 1, amended)

Report No. 1. A metrological analysis of belief in the proposition that consuming approved GM food is dangerous to human health, verified FALSE at VFRB tier 1 against the World Health Organization, the EFSA GMO Panel and the 2016 US National Academies review. The anchor measurement is Eurobarometer 73.1 (2010), on the split-ballot item base rather than the full wave. The principal finding concerns the widely cited fall in European concern about GM food between 2005 and 2019: that series spans an undocumented change of question format, the readings on the instrument used in 2005 and 2010 were essentially unchanged, and no part of the apparent decline can be attributed to belief change until the two formats are harmonised. An uncharacterised change in the measuring instrument was read as a change in the measurand. The uncertainty budget follows JCGM 100:2008 with Type A and Type B components combined in quadrature; the instrument-change term dominates the combined variance. Capability quantities follow Currie (1968) and ISO 11843-1 and are reported as properties of the procedure. No decision is issued against a zero-prevalence null, which cannot be prepared. This version supersedes the reading published 29 July 2026: the anchor item base is corrected to the split-ballot half-sample, two Eurobarometer waves are re-attributed, the anchor item is restated as disagreement with a positively worded safety statement, and the Type B evaluation is complete rather than provisional. The superseded figures are recorded rather than removed.