Key Comparisons and International Interlaboratory Studies Project

Interlaboratory studies have long been used to ensure measurement capability for commerce since accurate measurements are necessary for assessing product specifications.  For this reason design and analysis of interlaboratory studies have been an important part of the ITL Statistical Engineering Division's (ITL/SED) work for many years. Recently, however, a new type of interlaboratory study, known as a key comparison, has taken a critical new place in the NIST mission.  In the last year, key comparisons, international interlaboratory studies for comparing measurement results between leading National Metrology Institutes (NMI's), have provided many new opportunities for ITL/SED to collaborate with scientists across NIST. The impetus for these new opportunities is a Mutual Recognition Arrangement (MRA) signed by the NMI's belonging to the International Committee for Weights and Measures (CIPM) "to establish the degree of equivalence of national measurement standards maintained by NMI's, to provide for the mutual recognition of calibration and measurement certificates issued by NMI's, [and] thereby to provide governments and other parties with a secure technical foundation for wider agreements related to international trade, commerce and regulatory affairs."

Key comparisons serve as the technical basis for the MRA and must therefore accurately reflect the true relationships between measurement systems maintained by NMI's belonging to the CIPM. The results of key comparisons must also be extensible to members of Regional Metrology Organizations (RMO's) to maximize recognition of measurement capabilities that exist in other metrology laboratories around the world.

The contributions of collaborators include the following.

• Design of comparison studies.
• Coordination with BIPM and comparison participants.
• Data collection.
• Dissemination of Key Comparison results.

ITL/SED contributions to interlaboratory studies and key comparisons include the following.

• Performed data analyses which account for covariances in the measurements used to compute differences in results between labs and which ensure that uncertainties have a confidence level of 95%, as outlined in MRA policies.
• Contributed in the computation of reference values from interlaboratory data. Reference values are a sometimes controversial part of the MRA whose real role is still currently being worked out at both the policy and technical levels. The earliest key comparisons, those analyzed to date, have often used designs which allow for straight-forward computation of the necessary measurement differences but give less consideration to computation of uncertainties. Based on contacts from current work, however, ITL/SED has recently begun to contribute to comparison designs, ensuring that data collection will be as effective as possible. Opportunities to impact the design and analysis of comparisons have also been bolstered by participation in international meetings on interlaboratory studies organized by statisticians and other researchers from NMI's.