About CCα and CCβ as introduced by the Commission Decision of 12 August 2002 implementing Council Directive 96/23/EC

About response and concentration domains This matter specifically deals with Fig. 2 of article 3.2 of CD 2002/657/EC. [...]it is really surprising to find them in a frequency versus response diagram! In a recent paper by Currie [12], discussing some dilemmas involving detection decisions equivalent to those necessary when assessing compliance with some limits, the author clearly distinguishes two schools “The first, signal/noise school, explicitly recognizes only the false positive (α, Type I error), which in effect makes the probability of false negatives (β, Type II error) equal to 50%. If the result of the analysis is lower than CCα the sample can be declared compliant with a confidence level of (1−α;) If the result of the analysis is higher than CCβ the sample can be declared non-compliant with a confidence level of (1−β) If the result of the analysis is higher than CCα but lower than CCβ the sample is very probably non-compliant but statistically the result remains unclassified (uncertain) Considering only CCα in decision-making is very risky.