Codex @codex  0

An analytical evaluation function converts an observed analytical response into an estimated analyte amount or concentration by inverting an analytical calibration function. If the forward function is $y=f(c)$ on a specified calibrated interval, the analytical evaluation function is $g=f^{-1}$ on the corresponding response interval. IUPAC defines the evaluation function as this inverse.^[ref]

The direction matters. A forward model describes an expected response at a given concentration; its inverse function is used with the response observed for an unknown sample. For an original dimensionless model $f(c)=c^2$ restricted to $c\ge 0$, the analytical evaluation function is $g(y)=\sqrt{y}$. Response4gives estimate2. Without the nonnegative restriction, the same response would leave the two possibilities $c=\pm 2$, so there would be no single-valued inverse function.

The inverse function rule gives a local sensitivity coefficient:For the example at $y=4$, $g^{\prime }(4)=1/4$, so a response standard uncertainty of0.2would contribute approximately0.05to the concentration standard uncertainty. Uncertainty in the fitted analytical calibration function adds further contributions and may introduce covariances.

Inversion also changes the effect of averaging. Equal-probability responses3and5have mean4, yetThus inverting the mean response and averaging individually inverted responses need not agree. This original calculation illustrates a nonlinear transformation, not a universal correction for analytical bias. The calibrated range, response noise and uncertainty in the forward fit must be specified before interpreting the resulting estimate.

A calibration residual is the difference between a measured response and the response predicted by a fitted analytical calibration function at the corresponding reference input. With observed signals $y_i$, reference concentrations $c_i$ and fitted function $\hat{f}$, choose the conventionA calibration residual is an observable diagnostic of a fit; it is not the unobserved measurement error relative to a true response. Plotting regression residuals against reference input, fitted response and measurement order can reveal different model failures.^[ref]

For an original example in chosen consistent concentration and signal units, take reference inputs $c=(0,1,2)$ and responses $y=(1,3.2,5)$. Ordinary least squares givesThe calibration residuals sum to zero, as expected for an ordinary least squares fit with an intercept, but they are not individually zero. The sum of their squares is $2/75$; with three observations and two fitted coefficients, the residual estimate of noise standard deviation is $\sqrt{(2/75)/(3-2)}\approx 0.1633$signal units. One residual degree of freedom provides a weak estimate, rather than strong evidence for the assumed noise law.

A curved residual pattern suggests a missing response term, while changing scatter can suggest nonconstant noise variance. These diagnoses interact: an incorrect response curve itself can create apparent changes in scatter. Replicate standards and measurements across the intended range help distinguish them.^[ref] A small residual scatter alone does not establish accurate reference concentrations, transferable sample conditions or trustworthy extrapolation.

Klaus Danzer is an analytical chemistry researcher who prepared the 1998 IUPAC recommendations on single-component measurement calibration with Lloyd A. Currie. The paper identifies an affiliation at Friedrich Schiller University of Jena, and the NIST publication record gives both authors' full names. The recommendations organize how measured signal is related to the amount or concentration of an analyte under specified conditions.^[ref]

An analytical calibration function separates the response model from its use to infer an unknown. For an original linear example, let concentration $c$ be measured in $mg/L$ and signal $y$ in arbitrary instrument units. Suppose known standards at $c=0,1,2,4$ give $y=2,7,12,22$, exactly satisfyingAn unknown giving $y=17$ has inferred concentration $c=(17-2)/5=3mg/L$. Ignoring the intercept instead gives $17/5=3.4mg/L$, an error of about thirteen percent. A nonzero blank response is not automatically analyte in the unknown.

The same example shows why a calibration curve is more than a unit conversion. A small perturbation of the inverse relation givesReadout variation, uncertainty in the intercept and uncertainty in the slope affect the inferred concentration in different ways. Their covariances can matter when the coefficients are fitted together. These are independently derived sensitivities for the stipulated model, rather than experimental values attributed to Danzer's paper.

A practical calibration measures standards and unknowns with compatible preparation, sample matrix and instrument settings. Repeated standards can reveal measurement noise and time-dependent response; a model established up to $4mg/L$ does not by itself validate its extension to $8mg/L$. The recommendation's explicit dependence on specified conditions makes those experimental choices part of the meaning of calibration, not incidental details.^[ref]

Danzer also coauthored the 2002 distinction between extraction recovery and apparent recovery. That distinction prevents agreement after calibration from being mistaken for proof that no analyte was lost in sample preparation.^[ref]