Inferred values, accuracy

The accuracy of a method is defined as the closeness of the value obtained to known or accepted values. Accuracy can be determined in a number of ways, depending on the nature of the CZE method and availability of orthogonal techniques to compare results. If practical, spike recovery studies (i.e., testing to determine whether recovery matches the amount of a known analyte or impurity spiked) are good alternatives to orthogonal assay comparisons. ICH guidelines also allow method accuracy to be inferred, once specificity, linearity, and precision are established. 

In their broadest application, CRMs are used as controls to verify in a direct comparison the accuracy of the results of a particular measurement parallel with this verification, traceability may be demonstrated. Under conditions demonstrated to be equal for sample and CRM, agreement of results, e.g. as defined above, is proof. Since such possibilities for a direct comparison between samples and a CRM are rare, the user s claims for accuracy and traceability have to be made by inference. Naturally, the use of several CRMs of similar matrix but different analyte content will strengthen the user s inference. Even so, the user stiU has to assess and account for all uncertainties in this comparison of results. These imcertainty calculations must include beyond the common analytical uncertainty budget (i) a component that reflects material matrix effects, (2) a component that reflects differences in the amount of substance determined, (3) the uncertainty of the certified or reference value(s) used, and 4) the uncertainty of the comparison itself AU this information certainly supports the assertion of accuracy in relation to the CRM. However, the requirement of the imbroken chain of comparisons wiU not be formally fulfilled. 

Research by Burton and Price (38) demonstrated that Ba/Sr ratios generated by ICP emission spectroscopy (ICP-ES) can be used to infer the diet (marine verses terrestrial) of prehistoric populations. In this experiment we duplicate results obtained by Burton and Price for the Paloma samples. Our results show that the Ba/Sr ratios obtained by LA-ICP-MS are comparable in precision and accuracy to ICP-ES data (Figure 12). Although it is not unexpected that a coastal population would rely heavily upon marine resources, there are applications where this type of research would have value. What we have done here is demonstrate the efficacy of LA-ICP-MS to this line of research by demonstrating that it is possible to generate results similar to those obtained by other analytical techniques. 

In the absence of actual experimental data, we recommend method 1 below to estimate H or Kaw. We then recommend methods 2 and 3 as comparable in accuracy. It may be prudent to apply both methods. Finally (method 4), it may be possible to infer a value of Kaw from those reported for structurally similar compounds. Two of these methods are illustrated for four benchmark chemicals in the Appendix. 

The work of this laboratory extends the defect treatment to intermetallie compounds. The experiments measure simultaneously both the cadmium vapor pressure and the composition at equilibrium for a series of only slightly different alloy compositions. The precision and the relative accuracy of the measurements are high the absolute values suffer from any starting composition uncertainty and from errors in the absolute vapor pressure of cadmium as determined by other techniques. The experimental method is described elsewhere in this symposium (6). It has proved possible to infer the concentration and identity of lattice defects by analyzing the experimental data following the analytical techniques described below. 

We also need to mention something about the accuracy and precision of numbers as used in practice. Accuracy refers to how close a measured value is to its true value precision refers to the degree of dispersion (or deviations) of measurements from their true values. In this book you will encounter integers such as 1, 2, 3 and so on, which are in some cases exact (2 reactors, 3 input streams) but in other cases are shortcut substitutes for presumed measurements in problem solving (3 moles, 10 kg). You can assume that 10 kg infers a reasonable number of significant figures in relation to the other values of parameters stated in an example or problem, such as 10 kg o 10.00 kg. You will also occasionally encounter fractions such as which can also be treated as 0.6667 in relation to the accuracy of other values in a problem. 

The values of the properties which will be fitted by using equations 2.9 and 2.10 will be selected from available and apparently reliable experimental data whenever there are sufficient amounts of such data. Some important properties of polymers, such as the van der Waals volume (Chapter 3) and the cohesive energy (Chapter 5), are not directly observable. They are inferred indirectly, and often with poor accuracy, from directly observable properties such as molar volume (or equivalently density) and solubility behavior. When experimental data are unavailable or unreliable, the values of the properties to be fitted will be estimated by using group contributions. The predictive power of such correlations developed as direct extensions and generalizations of group contribution techniques will then be demonstrated by using them... 

Accuracy is about getting the right answer. It is an estimate of how close our measured value is to the true value. Knowing the true value can be very difficult but it is normally done by reference to recommended values for intemadonal geochemical reference standards (see for example Govindaraju, 1984 Abbey, 1989). It is of course possible to obtain precise, but inaccurate, results, For most geological studies precision is more important than small differences in absolute concentration, for provided the data have all been obtained in the same laboratory the relative differences in rock chemistry can be used to infer geochemical processes. 

A conceptually different and relatively new example of an inferential model, motivated by human performance problems specifically, is nonlinear causal resource analysis (NCRA) [Kondraske, 1988 Vasta and Kondraske, 1994]. Quantitative task demands, in terms of performance variables that characterize the involved subsystems, are inferred from a population data set that includes measures of subsystem performance resource availabilities (e.g., speed, accuracy, etc.) and overall performance on the task in question. This method is based on the following simple concept Consider a sample of 100 people, each with a known amount of cash (e.g., a fairly even distribution from 0 to 10,000). Each person is asked to try to purchase a specific computer, the cost of which is unknown. In the subgroup that was able to make the purchase (some would not have enough cash), the individual who had the least amount of cash provides the key clue. That amount of cash availability provides an estimate of the computer s cost (i.e., the unknown value). Thus, in human performance, demand is inferred from resource availabdities. 

It can he said that Winsor s H ratio has the best pedagogical value of ail formulation concepts because it is simple to understand and at the same time allows quite good qualitative predictions. If it were amenable to numerical calcu-laiii>iis there would be no need for alternate formulation concepts. This is mtt tfk case, however, hccniise the molecular interact ions Ajo cannot bt calciilaied with siifhcieiii accuracy yet. The R ratio would thus be the choice insirumcm to cany out qualitative inferences and to predicts trends. However, a quantitative... 

---