Regulatory authorities are currently discussing the measurement and imposition of ceilings on certain smoke constituents. However, if routine measurement of these constituents is to be required, the laboratories need to use the same validated methods and the precision of the methods has to be known. To assess method precision, the widely used method is to perform a collaborative study in order to determine the repeatability and reproducibility. Therefore, comparing data coming from different laboratories are relevant only if all the uncertainties are taken into account. For doing that, the ISO 5725 part 6 ^[1] recommends using the critical difference (described in the standard) that is based on the precision of the method. In this paper, using the CORESTA 2006 joint experiment data ^[2], we show, on a number of smoke constituents from reference cigarettes, the importance i) to use the appropriate statistical methods to compare results from different laboratories in order to avoid misleading conclusions; ii) to have validated and standardised methods with known precision. Moreover, using the critical difference computed from repeatability and reproducibility of the methods, we demonstrate that the number of replicates have a small effect on product comparison.

[1] ISO5725-6, 2001. Accuracy (trueness and precision) of measurement methods and results. Part 6. Use in practice of accuracy values
[2] M. Intorp, S. Purkis, M. Whittaker and W. Wright, 2009. Determination of Hoffmann Analytes in cigarette mainstream smoke. The CORESTA 2006 joint Experiment. Beiträge zur Tabakforschung Int. 23, 161-202