Compile a comprehensive list of A/B tests that deal with numerical data. This should include t-tests (independent, paired, one-sample), ANOVA, repeated measures ANOVA, etc. Since you mentioned having the most comprehensive list, we should also consider non-parametric equivalents where appropriate (Mann-Whitney U, Wilcoxon signed-rank, Kruskal-Wallis, etc.) since they can be applied when data do not meet the assumptions of parametric tests.
Compile a comprehensive list of A/B tests that deal with numerical data. This should include t-tests (independent, paired, one-sample), ANOVA, repeated measures ANOVA, etc. Since you mentioned having the most comprehensive list, we should also consider non-parametric equivalents where appropriate (Mann-Whitney U, Wilcoxon signed-rank, Kruskal-Wallis, etc.) since they can be applied when data do not meet the assumptions of parametric tests.