How the choice of data reduction can strongly influence uncertainty assessment: A re-analysis of Mn-bath experiments

The Mn-bath technique is widely used, especially by standardization laboratories, for the absolute determination of neutron emission rates. Understanding the limitations of the technique, and in particular the total measurement uncertainty, is crucial if quality results, fit for purpose, are to be reported. In this work, we show that the way in which the acquired data is analyzed can strongly influence the uncertainty assessment. We take a carefully performed set of Mn-bath measurements from the literature as our example and show that the same data when reanalyzed can be used to justify an uncertainty smaller by about an order of magnitude than was originally reported. This finding should caution all those involved in radiation measurements to critically assess their approach to data analysis and to perform a careful uncertainty analysis taking into account possible alternatives.

► Carefully assessing and reporting uncertainties in experimental work is vital.
► The uncertainty or bias introduced by the procedure adopted matters.
► Different analysis on the same data can give dramatically different uncertainty.
► We show a factor ten reduction in the standard deviation.