ASTM WK83525
New Practice for Equivalence testing for comparing test results
1. Scope
This practice provides statistical methodology for conducting equivalency testing on numerical data from two sources of test results to determine if their true mean, variances, or other parameters differ by no more than a published limit.Keywords
precision; sample size; quality determination; quality managementRationale
Quality management practices allow test results from different laboratories to be used for product acceptance and payment. Current statistical test procedures are used to determine whether statistical differences exist between data sources. These test procedures are sensitive to sample size and lead to disputes over data acceptance. Some data comparisons with a small variations (acceptable data variability) may be rejected because the calculated confidence interval doesn't include zero. Those data sets with a high degree of variability (unacceptable), are found to be statistically non-different as the calculated confidence interval includes zero and are accepted. This standard will be used by local, state and federal agencies, consultants, and contractors.
Work Item Status
Date Initiated: 09-21-2022
Technical Contact: Richard Steger
Item: 000
Ballot:
Status: