Replicate tests, determination appropriate number

Unfortunately it is not possible to keep a and 5 small at the same time. In practice it is common to specify a small a first, also called the testing level, and determine an appropriate number m of replications to keep 5 as small as possible afterwards. 

The measurement of triboelectric charge generation is well known for its variability where relative humidity, surface contamination or abrasion, tress configuration, and other variables have to be strictly controlled. Even with the combing of up to five replicate tresses and five successive determinations, Lunn and Evans report a 95% confidence interval of 15%. The apparatus that Lunn and Evans used is shown in Figure 28. The insulated tress is combed with a hand-held insulated test comb, and after the appropriate number of strokes, the tress is released into a Faraday cage connected to an electrometer. The whole setup is enclosed in a box in which temperature and humidity are controlled. 

Such threshold values are often estimated using no-observed-effect concentrations or levels (NOECs or NOELs). It might be tempting to substitute the individual ECx values in the CA equation (Equation 4.2) with NOELs in order to calculate a mixture NOEL. But this would imply that all NOELs provoke the same, statistically insignificant effect that is, all of them must have been determined in an identical experimental setup (in terms of number of replicates, spacing of test concentrations, variance structure), which is hardly ever the case. Nevertheless, a range of methods, such as TEFs or TEQs (see Chapters 1 and 5), makes use of a CA-like approach and sums up NOEL-based hazard quotients. This introduces an additional source of uncertainty in the risk assessment, which is fundamentally different from the question of whether CA is an appropriate concept for the mixture of interest. 

Establishment of Common Protocols and Standard Operating Procedures It is essential that all factors relevant to the conduct of the alternative method that may affect the results, the collection of data, and interpretation of the alternative method results be clearly defined before the study begins. These are best documented in the study protocol and SOPs that define the alternative methods. In order to assess the adequacy of the SOPs, they should be examined to determine if they contain three key elements. First, each SOP must have a detailed step-by-step description of how to conduct the assay. Enough details need to be provided such that any appropriately trained and competent laboratory technician need use only this document as the guide to run the assay. Second, the SOP must indicate the steps used to calculate the endpoint of the assay and the number of replicates necessary. Any data transformation or algorithms applied to the data should be clearly documented and consistently applied across all laboratories conducting a particular assay. Third, the protocol must specifically describe the prediction model being tested in the validation study. 

An appropriate sample size, or number of replicates, can be calculated for the type of statistical test by using the a and P error, the minimal detectable difference between two test procedures, and the variability (standard deviation) of data determined from previous neutralization system validations. The statistical test chosen to detect if there was a significant comparative increase or decrease in microorganism populations is the two-tailed, pooled Student s f-test. Both and values have been determined, 0.05 and 0.10, respectively. The minimal detectable difference is the minimal difference between samples from two procedures that the researcher would consider as significant and would want to be assured of detecting. Minimal differences that have been published are 0.15, 0.20, and 0.30 log 10 differences between data from Phase 1 and those from other phases [4,19,20]. The 0.15 logic difference will be used for this validation, because it is the most conservative and is from a validation test that involves multiple samples (replication) and a statistical analysis [4]. The final requirement, variability of the data, will be difficult to establish, especially because many researchers will be performing this validation for the first time. If past data are unavailable, then an option is to use an excessive sample size (at least 10) and use the data from that validation to determine an appropriate sample size for future validation studies. 

---