Close Figure 1: Conceptual framework for statistical analyses of data. Of the two kinds of variables, qualitative (categorical) and quantitative (numerical), qualitative variables (nominal or ordinal) are not normally distributed. Numerical data that come from normal distributions are analyzed using parametric tests, if not; the data are analyzed using non-parametric tests. The most popularly used Student's t-test compares the means of two populations, data for this test could be paired or unpaired. One-way analysis of variance (ANOVA) is used to compare the means of three or more independent populations that are normally distributed. Applying t test repeatedly in pair (multiple comparison), to compare the means of more than two populations, will increase the probability of type I error (false positive). In this case, for proper interpretation, we need to adjust the P values. Repeated measures ANOVA is used to compare the population means if more than two observations coming from same subject over time. The null hypothesis is rejected with a 'P' value of less than 0.05, and the difference in population means is considered to be statistically significant. Subsequently, appropriate post-hoc tests are used for pairwise comparisons of population means. Two-way or three-way ANOVA are considered if two (diet, dose) or three (diet, dose, strain) independent factors, respectively, are analyzed in an experiment (not described in the Figure). Categorical nominal unmatched variables (counts or frequencies) are analyzed by Chi-square test (not shown in the Figure) 