Sampling statistical analysis

Sampling. Statistical analysis relies on a clear distinction between random variability and variation in results due to deliberate manipulation of known factors in an experiment. Treatments are assigned systematically to subjects while individual subject characteristics contribute to random variability. It is important to understand that studies are not performed to learn what happened to the participants—they are conducted to provide a basis for predicting what will likely happen to an entire future subject population if a product is approved and broadly marketed. Inferences from a sample to a larger population are only possible when certain statistical principles are followed in the selection of that sample. Those principles are rarely followed in clinical research today, and inferences to future subject populations are therefore not generally supported from a statistical standpoint. 

Coliforms and enterococci are commonly used as indicators of unsanitary conditions in food processing. Clarke et al. [94] isolated enterococci from 18 out of 35 dehydrated vegetable samples. They found coliforms in 18 and enterococci and coliforms in 15 samples. Statistical analysis showed a positive correlation between number of enterococci and coliforms. The predominant species recovered from enterococci was Streptococcus faecium (60%) and from coliforms was Aerobacter (56%). 

The probabilistic nature of a confidence interval provides an opportunity to ask and answer questions comparing a sample s mean or variance to either the accepted values for its population or similar values obtained for other samples. For example, confidence intervals can be used to answer questions such as Does a newly developed method for the analysis of cholesterol in blood give results that are significantly different from those obtained when using a standard method or Is there a significant variation in the chemical composition of rainwater collected at different sites downwind from a coalburning utility plant In this section we introduce a general approach to the statistical analysis of data. Specific statistical methods of analysis are covered in Section 4F. 

A statistical analysis allows us to determine whether our results are significantly different from known values, or from values obtained by other analysts, by other methods of analysis, or for other samples. A f-test is used to compare mean values, and an F-test to compare precisions. Comparisons between two sets of data require an initial evaluation of whether the data... 

Vitha, M. F. Carr, P. W. A Laboratory Exercise in Statistical Analysis of Data, /. Chem. Educ. 1997, 74, 998-1000. Students determine the average weight of vitamin E pills using several different methods (one at a time, in sets of ten pills, and in sets of 100 pills). The data collected by the class are pooled together, plotted as histograms, and compared with results predicted by a normal distribution. The histograms and standard deviations for the pooled data also show the effect of sample size on the standard error of the mean. 

The most useful methods for quality assessment are those that are coordinated by the laboratory and that provide the analyst with immediate feedback about the system s state of statistical control. Internal methods of quality assessment included in this section are the analysis of duplicate samples, the analysis of blanks, the analysis of standard samples, and spike recoveries. 

Statistical errors of dynamic properties could be expressed by breaking a simulation up into multiple blocks, taking the average from each block, and using those values for statistical analysis. In principle, a block analysis of dynamic properties could be carried out in much the same way as that applied to a static average. However, the block lengths would have to be substantial to make a reasonably accurate estimate of the errors. This approach is based on the assumption that each block is an independent sample. 

If draws can be made from the posterior distribution for each component conditional on values for the others, i.e., fromp(Q,i y, 6,- J, then this conditional posterior distribution can be used as the proposal distribution. In this case, the probability in Eq. (23) is always 1, and all draws are accepted. This is referred to as Gibbs sampling and is the most common form of MCMC used in statistical analysis. 

The role of quality in reliability would seem obvious, and yet at times has been rather elusive. While it seems intuitively correct, it is difficult to measure. Since much of the equipment discussed in this book is built as a custom engineered product, the classic statistical methods do not readily apply. Even for the smaller, more standardized rotary units discussed in Chapter 4, the production runs are not high, keeping the sample size too small for a classical statistical analysis. Run adjustments are difficult if the run is complete before the data can be analyzed. However, modified methods have been developed that do provide useful statistical information. These data can be used to determine a machine tool s capability, which must be known for proper machine selection to match the required precision of a part. The information can also be used to test for continuous improvement in the work process. 

For many applications, quantitative band shape analysis is difficult to apply. Bands may be numerous or may overlap, the optical transmission properties of the film or host matrix may distort features, and features may be indistinct. If one can prepare samples of known properties and collect the FTIR spectra, then it is possible to produce a calibration matrix that can be used to assist in predicting these properties in unknown samples. Statistical, chemometric techniques, such as PLS (partial least-squares) and PCR (principle components of regression), may be applied to this matrix. Chemometric methods permit much larger segments of the spectra to be comprehended in developing an analysis model than is usually the case for simple band shape analyses. 

The water analysis is incomplete unless the number of coliform bacteria present is determined as well. A multiple-tube fermentation technique can be used to enumerate positive presumptive, confirmed, and fecal coliform tests. Results of the tests are expressed in terms of the most probable number (MPN). That is, the count is based on a statistical analysis of sets of tubes in a series of serial dilutions. MPN is related to a sample volume of 100 ml. Thus, an MPN of 10 means 10 coliforms per 100 ml of water. 

The bending of the graphite planes necessary to form a buckytube changes the band parameters. The relevant dimensionless parameter is the ratio a/R, where a ( = 3.4 A) is the lattice constant and R is the buckytube radius. For / = 20 A, the shift is expected to alter the nature of the conductivity[13-16j. In our buckybundle samples, most of material involves buckytubes with R > 100 A confirmed by statistical analysis of TEM data, and we assume that the elec-... 

In providing replicates for tests to be subjected to statistical analysis, it is necessary in the original sampling of the materials to be tested to ensure that normal variations in those qualities of the metals that might affect the results are represented in each set of samples. 

The sample meant here must not be confused -with a sample for chemical analysis. For a discussion of the sampling problem in statistics, see C. A. Bennett and N. L. Franklin, Statistical Analysis in Chemistry and the Chemical Industry, John Wiley and Sons, New York, 1954. 

The instrument has been evaluated by Luster, Whitman, and Fauth (Ref 20). They selected atomized Al, AP and NGu as materials for study that would be representative of proplnt ingredients. They found that only 2000 particles could be counted in 2 hours, a time arbitrarily chosen as feasible for control work. This number is not considered sufficient, as 18,000 particles are required for a 95% confidence level. Statistical analysis of results obtained for AP was impossible because of discrepancies In the data resulting from crystal growth and particle agglomeration. The sample of NGu could not be handled by the instrument because it consisted of a mixt of needles and chunky particles. They concluded that for dimensionally stable materials such as Al or carborundum, excellent agreement was found with other methods such as the Micromerograph or visual microscopic count. But because of the properties peculiar to AP and NGu, the Flying Spot Particle Resolver was not believed suitable for process control of these materials... 

All of these tests, by their nature, need to be repeated several times with different specimens of any polymer sample, in order to ensure that there is enough information for statistical analysis. Although the physicist Lord Rutherford said that if your results need statistics, you ought to have done a better experiment, his dictum cannot be extended to tests on mechanical... 

Besides, the statistical analysis of the results obtained confirmed that the xylan samples did not present a significant effect on the cell viability and cell proliferation rate when in direct contact with HeLa cells at the concentrations used in this study and compared to the control. 

Perhaps the most interesting aspect of this set of studies is the question posed in the recent paper by Schmidt et al. (2004) and deals with the reality of the patterns they observed. Is the polymorphism observed a result of the calculation methods used in the study, neural network (NN), and multivariate statistical analysis (MVA) Would increased sampling result in a greater number of chemo-types It is entirely possible, of course, that the numbers obtained in this study are a true reflection of the biosynthetic capacities of the plants studied. The authors concluded—and this is a point made elsewhere in this review—that ... for a correct interpretation a good knowledge of the biosynthetic background of the components is needed. ... 

This study of Phlox Carolina represents one of the best examples in the flavonoid chemosystematic literature where workers combined thorough sampling, detailed statistical analysis, and an intimate knowledge of the biology of the system under scrutiny to produce a convincing picture of natural variation. As mentioned at the beginning of this discussion, a more casual approach would have undoubtedly overlooked the subtle differences that characterize this system. 

Nilsson, T., Statistical analysis of individnal variation and sampling technique when determining the pigment content in beetroot, Swed. J. Agric. Res., 3, 201, 1973. 

Finally, a sample of 21 substances and the analysis of the standard deviations of measurements of LEL show that these standard deviations are not equal and therefore reflect different causes of variation. Without making the statistical analysis worse, the experimental values are very unstable eind therefore heirdly useful. 

---