Methods for Testing the Significance of Differences in Biological Testing

Abstract

Bioassays play a crucial role in the research of new drugs and vaccines. The significance test for differences is the most commonly used statistical method in the analysis of biometric data. The methods for testing the significance of differences in paired samples can be broadly divided into two categories. The first category is that the data population follows a normal distribution, and the commonly used method is paired t-test; The second type requires the use of nonparametric testing methods for statistical testing when parameter testing methods are not applicable. Commonly used methods include Wilcoxon matched signature level testing. This article introduces the basic principles and prerequisites of two types of testing methods, and in establishing a detection method for trypsin residue, takes the significance of the difference between the results of three enzyme-linked immunosorbent assay (ELISA) tests at 36℃ and the results at 37℃ for two batches of raw materials as an example to explain the correct use of the two testing methods. Correctly applying significance testing can establish the conclusions of experiments or investigations on a more scientific and reliable basis, avoiding simplification and absolutization.