ASTM D4483-05a(2011)

Tests are conducted using standard test methods to generate test data that are used to make decisions for commercial, technical, and scientific purposes. It follows that the precision of a particular test method is an important quality characteristic or figure of merit for a test method and a decision process.

An evaluation of the precision of a test method is normally conducted with (1) some selected group of materials as typically used with that method and (2) with a group of volunteer laboratories that have experience with the test method. The evaluation represents an event in time for the test method for these materials and laboratories. Another ITP precision evaluation with somewhat different materials or even with the same materials with the same laboratories at a different time, may generate precision results that differ from the initial ITP.

Experience as indicated in Refs (1-4) and elsewhere has shown that the poor reproducibility among the laboratories of a typical ITP is almost always due to interlaboratory bias. Certain laboratories are always low or high compared to a reference as well as other laboratories in all tests. This usual outcome for many ITPs is addressed in this practice by the use of the three-step robust analysis procedures as described in Section 7.

Caution is urged in applying precision results of a particular test method to product testing for consumer-producer product acceptance. Product acceptance procedures should be developed on the basis of precision data obtained in special programs that are specific to the commercial products and to the laboratories of the interested parties for this type of testing.

1.1 This practice covers guidelines for evaluating precision and serves as the governing practice for interlaboratory test programs (ITP) used to evaluate precision for test methods as used in the rubber manufacturing and the carbon black industries. This practice uses the basic one way analysis of variance calculation algorithms of Practice E691. Although bias is not evaluated in this practice, it is an essential concept in understanding precision evaluation.

1.2 This practice applies to test methods that have test results expressed in terms of a quantitative continuous variable. Although exceptions may occur, it is in general limited to test methods that are fully developed and in routine use in a number of laboratories.

1.3 Two precision evaluation methods are given that are described as robust statistical procedures that attempt to eliminate or substantially decrease the influence of outliers. The first is a General Precision procedure intended for all test methods in the rubber manufacturing industry, and the second is a specific variation of the general precision procedure designated as Special Precision, that applies to carbon black testing. Both of these procedures use the same uniform level experimental design and the Mandel h and k statistics to review the precision database for potential outliers. However, they use slight modifications in the procedure for rejecting incompatible data values as outliers. The Special Precision procedure is specific as to the number of replicates per database cell or material-laboratory combination.