We are receiving a lot of questions about relevant analyses in the Analyse-it Method Validation edition to help in evaluating new diagnostic tests in the fight against COVID-19. Below are some quick links that will help, but contact us if you have questions - we are working as normal.

- CLSI EP12-A2 for diagnostic accuracy see Sensitivity/Specificity, Predictive values, Estimating the sensitivity and specificity of a binary test and Comparing the accuracy of two binary diagnostic test.
- CLSI EP17-A2 for detection capability of molecular methods using Probit analysis, see: Estimating the detection limit using a Probit fit and may be useful for determining the cutoff of a method with an internal continuous response see: Estimating the detection limit of a measurement system and Estimating the detection limit of a measurement system using a precision profile.

Also see our latest blog post: Sensitivity/Specificity and The Importance of Predictive Values for a COVID-19 test

# Examples See what you can do with Analyse-it

Below are examples of all kinds of analyses using Analyse-it.

You can use them as templates to get started with your own analyses: each includes a dataset so you can see an example of how to arrange your data, and the analyses can be edited (if you have the appropriate edition of Analyse-it) so you can see how they are set-up.

The examples require Analyse-it version 4.80 or later.

#### All Analyse-it editions

Command | Features demonstrated | Download |
---|---|---|

Distribution | Speed of light distribution analysis. Demonstrates frequency histogram with custom classes, box and dot plot, CDF plot, Normal QQ plot, Shapiro-Wilk normality test, and 1-sample t-test. | Distribution 1.xlsx |

Compare Groups | Y by X analysis of independent groups. Demonstrates comparative box and mean-plots, 1-way ANOVA with Tukey-Kramer HSD multiple comparisons, mean-mean scatter plot, and Levene's test for homogeneity. | Compare Groups 1.xlsx |

Compare Groups | Calcium and Blood pressure analysis of effect of calcium vs. placebo on blood pressure. Demonstrates comparative box, mean and dot-plots, independent t-test, and F-test. | Compare Groups 2.xlsx |

Compare Pairs | Before / After study on the effect on bodyfat of an exercise regime. Demonstrates box and dot-plots, difference plot, Wilcoxon signed ranks test and Hodges-Lehmann estimator and confidence interval. | Compare Pairs.xlsx |

Compare Groups (contingency table) | Hair / Eye color analysis of independent groups frequency data (contingency table). Demonstrates clustered frequency plot, mosaic plot and Pearson X² (chi-square) test for independence. | Contingency Tables 1.xlsx |

Compare Pairs (contingency table) | Before / After intervention analysis of paired frequency data (contingency table). Demonstrates clustered frequency plot and McNemar test of difference between proportions. | Contingency Tables 2.xlsx |

Correlation | The Most Livable Neighborhoods in New York analysis of correlated variables. Demonstrates scatter plot matrix with density ellipses and frequency histograms, Pearson r correlation coefficient scatter matrix with confidence intervals. | Correlation.xlsx |

Fit Model | TV Advertising Yields 1983 regression analysis of retained impressions by advertising budget. Demonstrates Power function fit, scatter plot including fit and confidence bands, residual plot and residual diagnostic plots, and outlier / leverage / influence plot to identify important observations. | Fit Model 1.xlsx |

Fit Model | Pulse rates before and after exercise multiple regression analysis (including categorical predictor variables). Demonstrates advanced model fitting with dummy variables for categorical factors, partial residual leverage plots, residual plot and residual diagnostic plots, and outlier / leverage / influence plot to identify important observations. | Fit Model 2.xlsx |

Fit Model | Intensive care unit admissions logistic regression analysis of factors determining patient survival and discharge. Demonstrates logistic regression with multiple predictors, including tests of the effect of each term. | Fit Model 3.xlsx |

Compare Groups Fit Model |
Tensile strength of paper (Montgomery 2001) 1-way ANOVA analysis of strength by hardwood concentration. Demonstrates 1-way ANOVA using Compare Groups and the Fit Model command (either can be used, both techniques have their benefits). Includes model fit diagnostics, main effects plot, effect of terms, effect of means plot, and Tukey-Kramer multiple comparisons. | ANOVA 1.xlsx |

Fit Model | Aircraft primer paint (Montgomery 2001, page 572) 2-way ANOVA analysis of adhesion force by primer type and application method. Demonstrates model fitting with two factors and crossed/interaction term, main effect plots, and Tukey-Kramer HSD multiple comparisons. | ANOVA 2.xlsx |

Fit Model | Surface finish (Montgomery 2001, page 587) 3-way ANOVA analysis of surface finish. Categorical terms are coded as dummy variables. Demonstrates model fitting with multiple crossed terms, main effect plots, and 2- and 3-way interaction plots. | ANOVA 3.xlsx |

Fit Model | Viagra study (Field 2003) ANCOVA analysis of participant libido based on Viagra dose and partner's libido. Demonstrates ANCOVA model with covariate and factor variable, main effects plot, and Dunnett contrasts against placebo control group. | ANOVA 4.xlsx |

Multivariate | The Most Livable Neighborhoods in New York analysis of principal components. Demonstrates PCA analysis with coefficent matrix, and PCA bi-plot labelled with factors and locations with additional ideal location location point superimposed. | Multivariate.xlsx |

#### Analyse-it Quality Control and Improvement & Ultimate Editions only

Command | Features demonstrated | Download |
---|---|---|

Process Control | Copper concentration in plating pool process control analysis over 3 phases (production, IQ, and OQ). Demonstrates Shewhart X-bar R multi-phase control chart with 3-sigma control limits and color by "operator" , Montgomery rules to detect out of control points, control statistics for each phase, and EWMA and CUSUM control charts. | Control Example 1.xlsx |

Process Control | Loan processing costs (Montgomery page 268) analysis over 2 phases. Demonstrates Shewhart I-MR multi-phase control chart with 3-sigma control limits. | Control Example 2.xlsx |

Process Control | Supply chain operations (Montgomery page 325) error analysis from frequency data. Demonstrates Shewhart u (number of defects) control chart with 3-sigma control limits. | Control Example 3.xlsx |

Process Control | Purchase order data (Montgomery page 311) non-conforming unit analysis from frequency data with varying sample sizes. Demonstrates Shewhart p (proportion of defects) control chart with 3-sigma control limits and standardized Shewhart p-chart. | Control Example 4.xlsx |

Capability | Copper concentration in plating pool process capability analysis. Demonstrates Shewhart X-bar control chart with 3-sigma limits, process distribution histogram and box-plot, normality test and plot, capability statistics (Pp, Ppl, Ppu, Ppk), and performance against lower/upper specification limits. | Capability Example 1.xlsx |

Pareto | Downtime of colorimeter for measuring copper concentration Pareto analysis before and after training. Demonstrates Pareto plot for before and after training to identify most common causes, and Pareto analysis for before and after cross-stratified by operator. | Pareto Example 1.xlsx |

#### Analyse-it Method Validation & Ultimate Editions only

Command | Features demonstrated | Download |
---|---|---|

Method comparison | CLSI EP09-A3 - Appendix I example comparing two methods. Demonstrates partitioning the measuring interval into two ranges: 0 to 1.8, and 1.8 to 100 ug/L, then using a difference plot to determine average bias. | EP09-A3 Example 1.xlsx |

Method comparison | CLSI EP09-A3 - Appendix I example comparing two methods. Demonstrates Ordinary Least-Squares, Weighted Least-Squares, Deming, Weighted Deming and Passing-Bablok method comparison. | EP09-A3 Example 2.xlsx |

Method comparison | CLS EP21-A - Table 2 example comparing two methods. Demonstrates Bland-Altman difference plot with allowable difference bands, mountain plot, and calculating average bias with limits of agreement. | EP21-A Example 1.xlsx |

Method comparison | CLSI EP21-A - Table 3 example comparing two methods. Demonstrates Bland-Altman difference plot with allowable difference bands, mountain plot, and calculating average bias with limits of agreement. | EP21-A Example 2.xlsx |

MSA | CLSI EP05-A2 - Appendix B example of determining precision over 20-days, 2-runs, with 2 replicates. Demonstrates variability plot, and precision: repeatability, between run, within day, between day, and within laboratory. | EP05-A3 Example 1.xlsx |

MSA | CLSI EP05-A3 - Appendix B example of determining precision, over 6 samples, for 3 laboratories with 5-runs and 5 replicates. Demonstrates precision: repeatability, between run, within laboratory, between laboratory, and reproducibility, and precision profile using 3-parameter power variance fit function. | EP05-A3 Example 2.xlsx |

MSA | CLSI EP10-A3 - Appendix B example of preliminary evaluation of a procedure. Demonstrates linearity, precision, and trueness. | EP10-A3 Example 1.xlsx |

MSA | CLSI EP15-A3 - Table 8 example of user precision evaluation over 5-runs and 5 replicates. Demonstrates variability plots, precision estimates, and testing precision against claim. | EP15-A3 Example 1.xlsx |

MSA | CLSI EP15-A3 - Example 1Z example of user estimation of bias over 5 runs with 5 replicates. Demonstrates variability plot, bias estimate and confidence interval, and testing whether bias is significant. | EP15-A3 Example 2.xlsx |

MSA | CLSI EP06-A - Appendix C example of evaluating linearity over 5 diluted levels with 2 replicates. Demonstrates scatter plot showing fits, precision estimates, non-linearity at each level, and non-linearity plot with allowable non-linearity. | EP06-A Example 1.xlsx |

MSA | CLSI EP06-A - Appendix C examples of evaluating linearity for two methods over 5 diluted levels with 2 replicates. Demonstrates scatter plot showing fits, precision estimates, non-linearity at each level, and non-linearity plot with allowable non-linearity. | EP06-A Example 2.xlsx |

MSA | CLSI EP17-A2 - Appendix A examples of evaluating detection capability using blank and low-level samples. Demonstrates calculating limit of blank (LoB) using normal quantile method, limit of detection (LoD) using pooled SD of non-blanks, and bi-histogram pf distribution of blank vs non-blank material. | EP17-A2 Example 1.xlsx |

MSA | CLSI EP17-A2 - Appendix B example (generated data to achieve given mean/SD for each pool) of evaluating detection capability using precision profile of multiple non-blank samples. Demonstrates calculating limit of detection (LoD) from precision profile and known limit of blank (LoB). | EP17-A2 Example 2.xlsx |

MSA | CLSI EP17-A2 - Appendix D example (generated data to achieve given mean/SD for each pool) of evaluating detection capability using precision profile of multiple non-blank samples. Demonstrates estimating limit of quantiation (LoQ) using precision profile. | EP17-A2 Example 3.xlsx |

MSA | CLSI EP17-A2 - Appendix C example of evaluating detection capability using probit regression. | EP17-A2 Example 4.xlsx |

Diagnostic | CLSI EP28-A3C - Table 4 example of determining reference intervals for Calcium for Male and Female subjects. Demonstrates determining reference limits using non-parametric quantile method with Sex as a partitioning factor. | EP28-A3 Example 1.xlsx |

Diagnostic | CLSI EP28-A3C - Table 5 example of determining reference intervals for ALT (with lognormal distribution) for Male and Female subjects. Demonstrates transforming the data to a normal distribution then determining reference limits using non-parametric quantile method with Sex as a partitioning factor. | EP28-A3 Example 2.xlsx |

Diagnostic | CLSI EP28-A3C - example of transferring reference intervals for Mercury for Male and Female subjects to a new method. Demonstrates testing whether reference limits can be transferred from another laboratory. | EP28-A3 Example 3.xlsx |

Diagnostic | CLSI EP28-A3C - Appendix B example of determining reference interval for Calcium. Demonstrates determining reference limits using bi-weight robust method with Sex as a partitioning factor. | EP28-A3 Example 4.xlsx |

Diagnostic | CLSI EP24-A2 - Appendix D example of examining diagnostic performance (ability) of two tests. Demonstrates ROC curve analysis of a single test OxLDL with bi-histogram showing diseased/non-diseased subjects and decision threshold plot of sensitivity vs specificity. A second example demonstrates comparing ROC curves of OxLDL vs LDL, with ROC curve plot, area under curve, and curve comparisons (Delong et al). | EP24-A2 Example 1.xlsx |

Diagnostic | CLSI EP12-A2 - Example 10.3.1 example of comparing diagnostic performance of two paired qualitative tests for H. pylori. Demonstrates mosaic plots to visualise diagnoses, sensitivity/specificity and comparison of the difference between tests, plus calculation of predictive values. - | EP12-A2 Example 1.xlsx |

Diagnostic | CLSI EP12-A2 - Example 10.3.2 example of comparing two paired qualitative tests for H. pylori. Demonstrates PPA / NPA method comparison. | EP12-A2 Example 2.xlsx |