The reference value is the value against which your results are compared. In any proficiency test, the validity of the results depends on the quality and reliability of the established reference values.
NAPT's process for establishing reference values was designed in collaboration with leading statisticians in the metrology community, including technical oversight from statisticians at NIST. Using standard technical practices, a thorough analysis is performed to assign a reliable reference value and its associated uncertainty. We ensure that these values are stable, and suitable for meaningful comparison.
To ensure the credibility of these values, pivot and reference laboratories used in this process are accredited. In addition, we analyze multiple evaluations of the data set through the following statistical reviews, including: Two Sigma, Three Sigma, Chauvenet’s Criterion, Sample Median, Trimmed Mean, Interquartile Range, Q-Test, and Thompson Technique. These reviews are conducted to ensure sound and meaningful results.
Each proficiency test undergoes comprehensive technical reviews—before, during, and after distribution. These reviews are essential for ensuring test integrity, establish and/or validating reference data, and check trends or anomalies in the data. Our technical advisors are confident that NAPT’s review process is unmatched among ILC providers.
These requirements are documented in NAPT Quality Procedure 407-1: Data and Statistical Analysis Procedure. A reference value is assigned only after a comprehensive review of collected data. Prematurely assigning a value—such as relying on a single measurement—is not technically sound and undermines confidence in the data. No single laboratory is infallible. That is why NAPT prioritizes a full, statistically rigorous analysis before finalizing reference values.