Reproducibility and Repeatibility

Both measures refer to the random error introduced every time a given property of a sample is measured. The distinction between the two must be defined for the specific problem at hand. Examples for continuous (Fig. 1.6) and discrete (Fig. 1.7) records are presented. 

In mathematical terms, using the additivity of variances rule, 

Each of these variances is the square of the corresponding standard deviation and describes the effect of one factor on the uncertainty of the result. 

Freprod — Frepeal F Ftemp F Foperator + Fchemicals + Fwork-up F Fpopulation + (1 -6) 

---

Laboratory tests used in the development of inhibitors can be of various types and are often associated with a particular laboratory. Thus, in one case simple test specimens, either alone or as bimetallic couples, are immersed in inhibited solutions in a relatively simple apparatus, as illustrated in Fig. 19.34. Sometimes the test may involve heat transfer, and a simple test arrangement is shown in Fig. 19.35. Tests of these types have been described in the literatureHowever, national standards also exist for this type of test approach. BSl and ASTM documents describe laboratory test procedures and in some cases provide recommended pass or fail criteria (BS 5117 Part 2 Section 2.2 1985 BS 6580 1985 ASTM 01384 1987). Laboratory testing may involve a recirculating rig test in which the intention is to assess the performance of an inhibited coolant in the simulated flow conditions of an engine cooling system. Although test procedures have been developed (BS 5177 Part 2 Section 2.3 1985 ASTM 02570 1985), problems of reproducibility and repeatability exist, and it is difficult to quote numerical pass or fail criteria. 

The infralaboratory calibration study was performed by the Institute for Environmental Studies. Sediment was extracted and cleaned up as indicated here. The determination of dioxin and/or dioxin-like content was according to the method indicated under the section DR CALUX analysis. For the intralaboratory study, the following parameters were investigated limit of detection (LOD), limit of quantitation (LOQ), and reproducibility and repeatability of the bioassay. 

Diesel cetane number reproducibility concerns usually develop whenever motor cetane numbers differ by more than one number. The reproducibility variation arises due to operator and engine differences. The typical reproducibility variance accepted by ASTM will change with increasing cetane number. The following reproducibility and repeatability limits have been established by ASTM for method D-613. 

Current tests are generally destructive (i.e., sample is altered or destroyed) and robust estimates of measurement system variability (all aspects of the procedure including the operators) are difficult to obtain without using methods such as Gauge R R-reproducibility and repeatability (14). Suitability of current methods then is based on calibration using a calibrator system that has its own built-in variability and other assumptions (e.g., in physical testing such characteristics as size, shape, density can alter aerodynamic and/or hydrodynamic behavior of materials in a test system and contribute to systems variability). 

Calculated repeatability, intermediate precision, and reproducibility values can be compared with those of existing methods. If there are no methods with which to compare the precision parameters, theoretical relative reproducibility and repeatability standard deviations can be calculated from the Horwitz equation and the Horrat value (Table 5). Horwitz RSD values are reported in Table 6. Higher variability is expected as the analyte levels approach the detection limit (see below). Next to the Horwitz equation, the AOAC s Peer Verified Program proposes its own levels of acceptability of %RSD as a function of analyte concentration level [56,72]. 

Use a top-down approach using reproducibility and repeatability standard deviations from an interlaboratory study by the Harmonized IUPAC/AOAC protocol (Horwitz 1995) or ISO 5725 (ISO 1994). 

Determination of volatile matter content using a slower heating rate is applicable to a wider variety of coals. However, the values obtained are sometimes lower (1 to 3% absolute) than those obtained from the regular method. This illustrates the empirical nature of this test and the importance of strict adherence to detailed specifications. The complexity of the constituents of coal that undergo decomposition during this test makes it necessary to have wide tolerances for reproducibility and repeatability. 

NBS s involvment in SRM s for gas transmission studies arose out of activities within ASTM Committees D-20 (Plastics) and F-2 (Flexible Barrier Materials) that were aimed at standardizing a new instrumental method employing a coulometric oxygen detector for the measurement of oxygen gas transmission rates in materials used for various kinds of packaging (4 ). A subsequent reevaluation of the precision data for the classical manometric and volumetric methods (5) for gas transmission measurements showed that there are serious problems with the reproducibility and repeatability these methods. The use of SRM 1470 to verify the calibration is currently being written into ASTM standards D-3985 and D-1434. 

In general, results from investigations based on measurements may be falsified by three principal types of errors gross, systematic, and random errors. In most cases gross errors are easily detected and avoidable. Systematic errors (so-called determinate errors) affect the accuracy and therefore the proximity of an empirical (experimental) result to the true result, which difference is called bias. Random errors (so-called indeterminate errors) influence the precision of analytical results. Sometimes precision is used synonymously with reproducibility and repeatability. Note that these are different measures of precision, which, in turn, is not related to the true value. 

Remember that British Standard BS 5532 [CAULCUTT and BODDY, 1983] provides qualitative and quantitative definitions of both reproducibility and repeatability. The determination of repeatability and reproducibility for a standard test method by interlaboratory tests is given in [ISO 5725]. 

An important requirement for toxicity tests is their reproducibility and repeatability. In conformation with the principles of good laboratory practice (GLP),27 it is recommended that they be performed according to standard procedures and guidelines prepared by world standardization organizations such as the OECD, ISO, or CEN. Table 9.1 presents a list of selected ISO standards and OECD guidelines with regard to the modality of toxicity tests. 

When the food to which the patient is sensitized is removed from the diet for 2-4 weeks (depending on symptoms), reintroduction can induce the same clinical reaction (reproducibility and repeatability of symptoms) (Kaczmarski et al., 1997 Sampson, 1999a Sicherer and Sampson, 1999 Burks, 2000). 

This is to allow all of the active sites on the silica gel to acquire the correct charge and thns bind to the ions from the mobile phase. Where ion-exchange processes are used for HPLC separations, it is particnlarly important that the stationary phase is fuUy equilibrated with the mobile phase so that reproducible and repeatable results can be obtained. If the stationary phase is not fully equilibrated, the separation processes for each analysis will be different. 

Batch experiments were performed at 298 K via a standard volumetric method, using a 1.1 L glass reactor at atmospheric pressure, containing typically 0.5 g of adsorbent, polluted by liquid VOC injection, leading to an initial concentration of about 0.5 mmol.L. Equilibrium times were 1 and 2 hours respectively for Fau Y and Sil Z, after which the gas phase was sampled and analysed by chromatography (HP 5890II). Using the conventional assumption that the clay binder does not take part in the adsorption mechanism, data were reported for pure zeolite material. Reproducibility and repeatability of experimental data were checked by three different manipulators. 

In the validation process the ultimate aim is to secure that the test methods are good enough with respect to representativeness, reproducibility and repeatability. How much effort should be spent on validation must be decided on a case by case basis. If large economic values as well as considerable health, safety and environmental issues are involved, much more emphasis must be paid to the validation of the test methods. The frequency of use of the test method should also be considered when determining the extent of validation. The total consequences of wrong results are of course larger for methods in extensive use than for test methods used occasionally. 

As these results accumulate, giving an indication of the reproducibility of the method, the need may arise to revise the warning and action limits, but drastic revision should not be undertaken until the source of any gross discrepancy between results for reproducibility and repeatability has been defined and, if possible, eliminated (e.g., by work simplification or further training of staff). 

Measurements taken directly in solution are less affected by changing environmental conditions and much more reproducible and repeatable results are generally obtained. Damping effects due to the liquid, however, can reduce the sensitivity of the sensor. Also the behaviour of a crystal in solution may not be due directly to the effect of mass on the surface. The frequency response of a layered crystal in solution is dependent on many factors, such as the viscosity and temperature of the liquid, and the thickness and rigidity of the immobilised layers. 

---