Each test method has a limited precision. Even if a test is performed as carefully and as correctly as possible on a material which is as homogeneous as can be obtained, the test will still vary from one to another. A widely used measure of the variation of the test results from a test method is the standard deviation (σ). In an ASTM standard test, the standard deviation of the test method can be found in the Precision and Bias statement for the test. The "Blue Book," Form and Style for ASTM Standards, requires that all test methods include Precision and Bias statements. Practice C 670 and Practice C 802 provide guidance for determination of these values.

If the precision of a test method is low and the precision of the test has not been properly considered in a material specification, a uniform material with the right quality may still be rejected most of the time because of the wide variation of the test results. In order to have rational specification limits, the precision of the test used should be properly included in a specification.

This practice provides a guideline for proper inclusion of precision of the test method in a rational specification.