Roadmap for outlier detection in univariate linear calibration in analytical chemistry: Tutorial review

Abstract Assessment of the adequacy of a proposed linear calibration curve is necessarily subjective in chemical analysis. If the outlier points in calibration are not identified and discarded, the constructed model will not have much validity and does not warrant the accuracy and precision of prediction step. Recognizing of influential points, outlier data, and discarding them is one of the steps in data processing that has been considered in various sciences. The outlier points can arise from (I) bad design of calibration set and (II) gross error in doing experiments. Therefore, we aimed to extract a map that recognizes the following issues: (A) the existence of data with high regression coverage that is far from the rest and will strongly affect the accuracy of the calibration equation, high leverage points; (B) large error in the experimental process: The recorded signal does not match the desired concentration; and (C) points with a concentration lower than the limit of quantification, which is calculated by considering the standard error of regression instead of the standard deviation of blank. The efficiency of the proposed roadmap will be reviewed, and this will give a new perspective on the calibration equation to avoid common mistakes in analytical chemistry. To achieve the above goal, visual and statistically significant tests will be used and all tests will be performed in a simple Microsoft Excel environment.