Comparing two analytical methods: minimal standards in forensic toxicology derived from information theory

Frequently, new instrumentation and techniques are being applied to casework without an adequate knowledge of how this new technology compares to acceptable analytical procedures. Information theory provides a mathematical method to estimate the identification power of various analytical procedures and predicts if the new method produces a better or poorer confidence in the analysis than an accepted method. This paper makes comparisons based on relative informational power and discusses methods to make those estimates. With the procedures outlined in this paper, an analyst can estimate if a novel technique has the opportunity of reaching the accepted analytical methodology and, if necessary, identify places where the greatest improvement can be made to reach this equilivalence. Even though relative information theory can dismiss some proposed techniques before any analysis starts, it cannot verify if a given technique is adequate for a particular task. For that purpose, a complete procedural verification must be undertaken. (C) 2004 Elsevier Ireland Ltd. All rights reserved.