Statistical Tools and Approaches to Validate Research Results

The quality of research depends on reliable research and its honest description. The search for alternative explanations and self-criticism is urgent. All methods have limitations, and the research involves multiple interpretations and a moral and ethical component inherent in judgments. The most important thing is to define the criteria to test the method's reliability and suitability for the intended use. It is also required to know whether the measurement result can be accepted with confidence or, on the contrary, rejected.

Credibility, authenticity, criticality, and integrity are the main criteria to validate the qualitative method. Selectivity, sensitivity, precision, and trueness are the most commonly defined criteria for describing the quantitive method.

The statistical tools allow us to address all these points and avoid personal bias. The experimental protocol can be based on a standard methodology inspired by regulatory guidelines regarding statistical data analysis in analytical method calibration and validation to optimize the number of assays and satisfy the study of validation criteria

In order to better understand the statistical analysis of raw data, simple, practical examples will be exercised for qualification and quantification method: the number of tests required to obtain reliable results (Balley's theorem, Student's t-distribution, and normal distribution of the results), the representation of the results (the mean of a data, standard deviation, and confidence interval), why we prefer to obtain a linear range for the quantitive method (matrix and other effects), random or statistically significant changes (superiority of method variance and signal to noise ratio).