Mandel, John (1972, ASQC) Institute for Materials Research, National Bureau of Standards, Washington, DC
A simple method is presented for the evaluation of repeatability and reproducibility. The method obviates such difficulties as transformation of data or weighted analysis of variance. The shortcomings of the method consist not in the estimates it gives for the two measures of precision, both of which are as good or better than those given by more sophisticated methods of analysis, but rather in that these two parameters do not contain all the pertinent information contained in interlaboratory test data. Also discussed is the rationale for a proper choice of the number of laboratories and replicate measurements required in an interlaboratory study.
Variation,Statistical tests,Statistics,Laboratories,Repeatability and reproducibility studies (R&R)