DescriptionParticular attention should be given to the concepts of precision, accuracy, rejection of outliers, standard addition, limit of detection (LOD), lower-limit of analysis (LLA) and upper-limit of analysis (ULA) where pitfalls in data management may lead to major errors in analysis. Several schools of QA promote different recommendations which may lead to inconsistent results and decisions. Therefore are required further investigations into the origin of significant deviations, which are observed for average values and uncertainties of e.g. inter-laboratory comparisons. Tentative correlation between results of inter-laboratory comparisons and intra-laboratory comparisons is discussed. Quality assurance (QA) and quality control (QC) are important tools of analytical chemistry, and several topics and concepts within statistics are required in order to perform reliable data management. Linear regression is applied to operational calibrations of analytical apparatus, and straight lines with correlation coefficients, most frequently, close to one are published in analytical chemistry. Calculation of regression line is a straightforward mathematical procedure, but the associated calculation of uncertainty needs more attention. Conversely, the standard-addition method (SAM) seems uncomplicated but there are a number of pittfalls that need be considered before reliable results may be produced.
Teaching in principles is important to students and the vast majority of time in the lecture theatre is spent on understanding of principles of chemical mechanisms. However when principles are supposed to be implemented in practical examples, that are important to innovation and entrepreneurship, then occupies estimation of uncertainty a key position; uncertainties lead the way to genuine results of science. Since the evaluation of statistical data has not yet arrived at a standardized stage of utilization in science, are treated advantages and drawbacks with respect to different methodologies of QA and QC. A profound understanding of uncertainties of linear regression is presented, which includes comparison of different methodologies and aiming at simplicity in data interpretation.
Invited speaker chaired by Reiner Salzer
|Period||26 Aug 2013|
|Event title||Euroanalysis XVII: Analytical Chemistry for Human Well-being and Sustainable Development|
- quality assurance
- quality control
- scientific methodology
- standard addition