Statistics in Analytical Chemistry

Abstract

Statistical techniques can be of considerable value in interpreting the results of experimental programs. Variation in experimental data may be due to either controllable or non-controllable factors in the experiment or in the methods of analysis used. The purpose of statistics is to separate and identify the controllable portion of the variation and to estimate the extent of the non-controllable portion. The key statistical concepts for evaluation of an analytical method are accuracy and precision. Precision can be subdivided into repeatability and reproducibility. Once these concepts are understood, confidence limits for a result or set of results can be obtained. All of these depend on our ability to measure the variation inherent in the method. Quick methods are available for estimating inherent variation and for detecting wild or out-of-control test results. Other methods distinguish between absolute and relative error and also show how error (or variation) propagates through a series of steps in the analytical procedure. Experimental design and analysis of variance are statistical methods which permit study of the sources of variation in a test procedure so that the controlable portion of the variation can be identified and eliminated.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 154 citations exist upstream.