To serve as a measurement standard, a (certified) reference material must b
e stable. For this purpose, the material should undergo stability testing a
fter it has been prepared. This paper looks at the statistical aspects of s
tability testing. Essentially, these studies can be described with analysis
of variance statistics, including variant regression analysis. The latter
is used in practice for both trend analysis and for the development of expr
essions for extrapolations. Extrapolation of stability data is briefly touc
hed upon, as far as the combined standard uncertainty of the reference mate
rial is concerned. There are different options to validate the extrapolatio
ns made from initial stability studies, and some of them might influence th
e uncertainty of the reference material and/or the shelf-life. The latter i
s the more commonly observed consequence of what is called 'stability monit
oring'.