IN PURSUIT OF ACCURACY - NOMENCLATURE, ASSUMPTIONS, AND STANDARDS

Accuracy is the central measure of quality in Analytical Science. Three topics, essential in the pursuit of accuracy, especially in the exposure and reduction of the "big" errors in analytical measurements, are reviewed in this article. The first topic, Analytical Nomenclature, lies at the heart of any effort to realize and communicate the nature and level of accuracy of the Chemical Measurement Process (CMP). The discussion centers about the historic role of IUPAC in this endeavor, together with some outstanding problems in Analytical Nomenclature, particularly as related to detection and identification. The second topic comprises assumptions and standards (materials and data) that are crucial in the search for major errors and in the control of accuracy, both within and between laboratories. Standard (Certified) Reference Materials have long been central to this effort, but more recently they have been joined by Standard Test Data. The latter, data sets having known characteristics, designed to simulate the structure of complex analytical signals, show great promise for the control of quality in the computational phase of the CMP. The final topic relates to Revolutions in Analytical Measurement Science that can make the most profound impact on accuracy: revolutions in sampling, measurement, and computation. Two illustrations are presented: (1) Accelerator Mass Spectrometry, a fundamentally new approach to atomic and isotopic mass spectrometry that makes possible the direct measurement of element and isotope ratios as small as 10(-16); and (2) Exploratory Statistical Graphics, a data visualization tool that permits analytical scientists to directly apply their intuitive "vision" to the assessment of multivariate data quality, and to search for unsuspected relationships in complex chemical datasets.