Statistical packages, using

Equations used to calculate percent yield or dilution ratios A list of disposable equipment (e.g., rubber gloves, Bunsen burners) Step-by-step instructions of the procedure Warnings to other scientists about unusual hazards Quantitative statements of reaction times and temperatures Descriptions of the physical appearances of synthesis products IR or NMR data confirming product purity Statistical packages used (including the name of the software) Reports of other software used to keep track of data (e.g., Excel)... 

The accuracy of a Hansch equation may be assessed from the values of the standard deviation (s) and the regression constant (r) given by the statistical package used to obtain the equation. The smaller the value of s the better the data fits the equation. Values of r that are significantly lower than 0.9 indicate that either unsuitable parameter(s) were used to derive the equation or there is no relationship between the compounds used and their activity. This suggests that the mechanisms by which these compounds act are unrelated because the mechanisms are very different from each other. 

The rainfall and toxin data will be entered into two appropriately labelled columns. You will then have to indicate the relevant columns. However, there is an important difference from correlation. With regression you must be careful to indicate correctly which is the dependent and which the independent variable. Unfortunately, statistical packages use a varied terminology. The toxin concentrations may be entered as the dependent variable or response and the rainfall may be the independent variable or the predictor . 

All calculations reported here were made by using the Time Series Modules of the MINITAB Statistical Package, running on a Data General MV8000 computer under the AOS/VS operating system. 

All samples were dried for 72 hours at 80°C, and dry weights were calculated. Dried samples were milled to spectrometrically measure the specific activity of i Cs. The standard error of specific activity was in the range 10-20%. Statistical analysis used the software package MS Excel. 

Analysis of variance (ANOVA) analyses were performed using the general statistical package StatView 5.01 (SAS Institute, Cary, NC, USA). The ANOVAs were calculated as repeated-measures ANOVAs with wells as within factor for phase 1 and with plates as within factor for subsequent phases. Specialized statistics, such as comparison of fits of different calibration curves, were calculated in MATLAB 5.1 (MathWorks, Natick, MA, USA) using custom routines. 

Statistical Methods. The Statistical Package for the Social Sciences (SPSS), version 8 (10, 11) was used for all statistical analyses. All data were entered into a permanent file and verified prior to statistical analyses. 

The SPSS statistical package for Windows was used [28], with level of significance was set at 5%. 

Quantitative methodology uses large or relatively large samples of subjects (as a rule students) and tests or questionnaires to which the subjects answer. Results are treated by statistical analysis, by means of a variety of parametric methods (when we have continuous data at the interval or at the ratio scale) or nonparametric methods (when we have categorical data at the nominal or at the ordinal scale) (30). Data are usually treated by standard commercial statistical packages. Tests and questionnaires have to satisfy the criteria for content and construct validity (this is analogous to lack of systematic errors in measurement), and for reliability (this controls for random errors) (31). 

Once the toxicity parameters were computed to a spreadsheet yielding a table of 30 rows (effluents) and 9 columns (bioassays), we ran a principal component analysis (PCA) to check the diversity patterns of effluents and the correlation between tests. The PCA calculations were carried out using the ADE 3.6 statistical package on a Macintosh computer. ADE was developed by the University of Lyon II and by the French National Centre of Scientific Research (CNRS) common biometry laboratory. The new version ADE version 4 running on Mac and PC computers is now available on this university s internet site at http //pbil.univ-lvon 1. fr/ADE-4/... 

Several statistics for multivariate tests are known from the literature [AHRENS and LAUTER, 1981 FAHRMEIR and HAMERLE, 1984] the user of statistical packages may find several of them implemented and will rely on their performing correctly. Other, different, tests for separation of groups are used to determine the most discriminating results in discriminant analysis with feature reduction. 

Calibration curves were constructed with the NIST albumin (5 concentrations in triplicate) and with the FLUKA albumin (5 concentrations in duplicate) in the concentration range of 50 250 mg/1. The measured values of individual concentrations fluctuated around the fitted lines, with a standard error of 0.007 of the measured absorbance. The difference between FLUKA and NIST albumin calibration lines was statistically insignificant, as evaluated by the t-test P=0.14 > 0.05. The calibration lines differed only in the range of a random error. The FLUKA albumin was, thus, equivalent to that of NIST. Statistical evaluation was carried out using the regression analysis module of the statistical package SPSS, version 4.0. 

Statistical Evaluation. Statistical analysis was performed on a Mini 6 Bull computer. The program package used was SPAD, Systeme Portable pour l Analyse des Donnees (15). Data from the chemical analyses were evaluated by principal component analysis (PCA) of correlation matrices. PCA was carried out in order to show clearly the association between chemical classes, or compounds (variables), and the isolates studied (individuals). Statistical analysis was also made on the basis of reduced chemical data, by calculating RV coefficients for single variables (17, 18). The number of variables, i.e. chemical compounds in this study, was reduced since only chemical compounds with the highest RV values were selected as representatives of the chemical group. 

With the exception of the EPISUITE package, which is freely downloadable, all the products listed in Table 3.7 are commercial and charges are associated with their use. Most enable large numbers of compounds to be input at one time, using most common file formats such as SMILES,. mol, and. pdb. Descriptors may be extracted easily from most of these packages and transferred into spreadsheets for statistical analysis. Some of the products include some form of statistical analysis, although the use of dedicated external statistical packages is recommended in most cases. 

The statistical method used to develop the QSAR should be specified (including details of any software packages used). 

There are many topics (such as searching chemical literature, Current Contents, or Citation Index the choice of hardware and maintaining the system use of statistical packages, etc.) that could be included into the book, but at some point a selection of what are the most common, and at the same time the most important, aspects of PCs in chemistry must be made. After all, the readers will judge if too many vital topics were omitted or not. 

An effective but simple way of graphically illustrating the variability associated with the analytical data is to plot x—y plots of the duplicate and replicate pairs. Most statistical packages will have an option for plotting simple x—y plots. The G-BASE project uses MS Excel running a macro that will automatically plot duplicate-replicate and duplicate-duplicate results. Figure 5.8 shows three examples from the G-BASE East Midlands atlas area duplicate-replicate data for soils. This method gives an immediate visual appreciation of any errors present in an analytical batch and an indication of within site variability, as shown by the duplicate pairs, or the within sample variability, as indicated by the replicate pairs that demonstrate... 

The G-BASE project has used several statistical packages to perform this nested ANOVA analysis (e.g., Minitab and SAS). It currently uses an MS Excel procedure with a macro based on the equations described by Sinclair (1983) in which the ANOVA is performed on results converted to logio (Johnson, 2002). Ramsey et al. (1992) suggest that the combined analytical and sampling variance should not exceed 20% of the total variance with the analytical variance ideally being <4%. 

Detailed instructions are provided for the calculation of the mean, median and SD (but not quartiles) using Microsoft Excel. Readers are referred to the accompanying web site for detailed instructions on generating all these descriptive statistics (including quartiles) using Minitab or SPSS. Generalized instructions that should be relevant to most statistical packages are provided in the book. 

---