Determination of moisture, solids-not-fat and fat-by-difference in butter using routine methods according to ISO 8851/IDF 191—an international collaborative study and a meta-analysis

An international collaborative study of routine methods for the determination of moisture, solids-not-fat (SNF) and fat-by-difference (i.e. fat (g 100g–1 product) =100−moisture (g 100g–1 product)−SNF (g 100g–1 product)) in butter was carried out to obtain estimates of the precision parameters repeatability (2.8sr) and reproducibility (2.8sR). Ten laboratories tested blind duplicate samples of eight different butters according to standard operating procedures jointly drafted by the International Standards Organisation (ISO) and the International Dairy Federation (IDF) (draft ISO standard 8851, parts 1–3; draft IDF standard 191, parts 1–3). Estimates of repeatability and reproducibility respectively were: moisture, 0.28g 100g–1 product and 0.44g 100g–1 product; SNF, 0.20g 100g–1 product and 0.34g 100g–1 product; fat-by-difference, 0.30g 100g–1 product and 0.51g 100g–1 product. These estimates are similar to those found previously in a New Zealand study. A meta-analysis, pooling the results of the present study and the New Zealand study, yielded overall estimates of repeatability and reproducibility as follows: moisture, 0.31g 100g–1 product and 0.42g 100g–1 product; SNF, 0.20g 100 g–1 product and 0.39g 100g–1 product; fat-by-difference, 0.35g 100g–1 product and 0.54g 100g–1 product, and these are recommended for adoption in the respective methods. This study completes the validation work for the respective methods, which may now be progressed for final adoption as international standard routine methods. A protocol is proposed for dealing with situations where multiple precision estimates are available.