Evaluation measurement

Intrusiveness. Workers are likely to alter their behavior, consciously or unconsciously, when they are observed. To the extent that a worker s exposure is related to the worker s actions, this change can distort the representativeness of the evaluation. Measurement methods which require the close presence of the person collecting the sample are more likely to influence the result than samples collected with unobtmsive devices worn by the worker. 

This chapter deals with handling the data generated by analytical methods. The first section describes the key statistical parameters used to summarize and describe data sets. These parameters are important, as they are essential for many of the quality assurance activities described in this book. It is impossible to carry out effective method validation, evaluate measurement uncertainty, construct and interpret control charts or evaluate the data from proficiency testing schemes without some knowledge of basic statistics. This chapter also describes the use of control charts in monitoring the performance of measurements over a period of time. Finally, the concept of measurement uncertainty is introduced. The importance of evaluating uncertainty is explained and a systematic approach to evaluating uncertainty is described. 

The concepts of precision, bias and accuracy were introduced in Chapter 4. However, as they are important in the context of evaluating measurement uncertainty it is worth revisiting them. 

It is important to have some knowledge of the reliability of all measurement results. Measurement uncertainty is the parameter used to describe the range within which the true value (or right answer) for a particular measurement is expected to lie. Evaluating measurement uncertainty involves a number of distinct steps, which are described in this chapter. 

In Section 4.2, we treated the performance of regression models. For classification methods, these concepts remain valid and can be directly used. However, there is an important difference concerning the performance measures to be used. While for regression the basic information for the evaluation measures are the residuals, i.e., the difference between observed and predicted v-values, this would... 

When using a model with two PLS components for the phenyl data, the thresholds corresponding to 95% are. yi. w 0.22 and yHiGH = 015, respectively. In this way, 36 out of 300 objects from the calibration set are not classified to any of the groups, corresponding to 12.0%. Using the same thresholds for the test set, 37 out of 300 objects (12.3%) are not classified. The evaluation measures for the classified objects are shown in Table 5.5, and they clearly improve compared to Table 5.4. 

In like vein, we would recommend the use of more than just one evaluation measure. That said, it is our experience that different measures usually agree as to the relative merits of different approaches (unless there are only very minor differences in effectiveness) even so, it is always worth carrying out additional analyses to ensure that one s results are, indeed, independent of the evaluation criterion that has been adopted. 

However, Fraser et al. (1998) noted that this canister sampling technique may underestimate the methyl-naphthalenes. Thus, their methylnaphthalenes/naph-thalene ratios were lower than those obtained by Arey et al. (1989a) using Tenax-GC solid adsorbent. This may be due to significant adsorption of the methyl isomers to the canister (Arey, personal communication). Zielinska and co-workers (1996) evaluated measurement methods for VOCs up to C2() emitted from motor vehicles and reported that C8-Cl2 hydrocarbons were more stable on the Tenax cartridge than in canisters. Similar problems with canister sampling for organics are discussed in Chapter ll.A.4e. 

The actual measurement of the refractive index of the solution poses no difficulty, but the evaluation of the refractive index gradient is more troublesome. The assumptions of the derivation of Equation (23) restrict its applicability to dilute solutions. The refractive index of a dilute solution changes very gradually with concentration hence a plot of n versus c, the slope of which equals dn/dc, will be nearly horizontal. Since the intensity ratio depends on the square of dn/dc, it is clear that successful interpretation of Equation (23) depends on the accuracy with which this small quantity is evaluated. Measuring the absolute refractive indices of various solutions and determining dn/dc by difference or graphically would introduce an unacceptable error. A more precise method must be used to measure this quantity. 

The Mollier I-X-Diagram for Evaluating Measurements on Cooling Towers... 

This way of evaluating measurement uncertainty, starting from the potential sources of error, was very useful in identifying those components that have a large contribution in the overall uncertainty and in minimizing them as much as possible. Also, good agreement... 

Evaluation measuring the effectiveness of proposed measures after... 

A somewhat less consistent observation is that the STG volume loss is significantly greater in the left hemisphere compared with the right. This inconsistency might be partially attributed to the part of the STG that was evaluated. Measurements restricted to the posterior STG seem to more commonly find a left > right... 

The enzymatic activity of a-chymotiypsin was evaluated in composites of poly(methyl metacrilate) with different carbon materials [91] demonstrating that the incorporation of SWCNTs into enzyme-polymer composites results in active and stable polymeric films. The release of the protein from the composite was evaluated measuring the enzymatic activity in the supernatant in contact with the composite. The results showed that in the case of SWCNTs the leaching of the protein from the composite was lower. This fact was attributed to the imion of the protein to the CNTs. The effect of other polymers such as polystyrene and poly(lactic acid) was also analyzed and the leaching of the protein was significant in the absence of SWCNTs. Only the hydrophobic ones (poly(methyl metacrilate)... 

A summary of the experimental results for fuel staging with different fuel combinations is presented in Fig. 9 The measured NO, emissions are plotted as a function of the temperature in the reburn zone for different reburn fuel rates. The CO emissions of the evaluated measurements were always below 500 mg/Nm at 11 % O2, usually below 250 mg/Nm ... 

The membrane characterization data reported in this section have been obtained by means of a home-made apparatus which is made of stainless steel and can operate from high vacuum up to 70 bars [17], It is characterized by the unique capability of performing a broad range of porous membrane characterization and evaluation measurements, namely equilibrium isotherms, absolute (integral and differential) and relative gas and condensed vapor permeabilities and selectivities. 

Experimental. IR measurements were done in a Beckman IR-5 spectrophotometer either on films or solutions. Films were prepared from 10% w/v or CH2CI2 polymer solutions. Solution measurements were done in a 0.01 mm path length NaCI cell. THF, DMF or CHClg were the solvents normally used. Near infrared spectra were recorded from a Beckman DK A spectrophotometer. 10 mm path length silica cells were used. Relatively high concentrations (5% w/v) of polymer in CHClg were used in this preliminary evaluation. Measurements were done at room temperature in all cases. Table II describes the copolymers used in the evaluation of the IR region. 

Klee GG. Cobalamin and folate evaluation measurement of methylmalonic acid and homocysteine vs vitamin B12 and folate. Clin Chem 2000 46 1277-83. 

Recommendation 4. The quantitative risk assessment (QRA) for each chemical demilitarization site should be iterative. Actual chemical events should be used routinely to test the completeness of the QRA, which should be routinely utilized to hypothesize the frequency and consequences of chemical events. The Program Manager for Chemical Demilitarization and the U.S. Army Soldier and Biological Chemical Command should use the QRAs to evaluate measures to control future chemical events. The Army should also consider using QRAs to examine scenarios associated with sabotage, terrorism, and war. 

The Davies-Bouldin validity index (DBI) is an example of a cluster evaluation measure (95). The DBI is the average similarity between each cluster and its most similar one. It is defined as... 

Test-retest reliability refers to the relationship between scores obtained from the same instrument on two or more separate occasions when all pertinent conditions remain relatively unchanged. It is usually evaluated using the intraclass correlation coefficient (ICC). However, QOL is not assumed to be constant over the course of time. In fact, most clinical studies attempt to assess how QOL changes. Test-retest reliability estimates may have limited value in evaluating measures that are designed to assess a dynamic process. 

The surface active properties were evaluated/measured by changes In the contact angle of the water droplet placed on the alr-slde surface and glass-side surface of the films. For the comparison of the efficiency of the surface modification, the results of random copolymers and homopolymers are also shown in Figure 1. An examination of the results shows that the slloxane polymers exhibit pronounced surface accumulation on that side. It can be seen that quite small amounts of slloxane polymers, e.g. 0.5-... 

Table 9.1 Contingency tabfe forming the base of evaluation measures...

The [i parameter regulates the respective influence of precision and recall. It is often set to 1 to give the same weight to these two evaluation measures. Back to our running example, using a /I equal to 1, the f-measure values obtained for COMA++ and SF are, respectively, as follows ... 

Practical Problems. As we have seen, many different types of disturbances cause changes in ecosystems that are similar and in broad outline predictable. What has been called (8) the strategy of ecosystem development provides a most important basis to anticipate the ecological effects of pollution and to evaluate measures of water quality control. Table II lists some examples of measures and how they grossly affect water quality. This brief list illustrates that water pollution control... 

---