Strictly Standardized Mean Difference, Standardized Mean Difference and Classical t-test for the Comparison of Two Groups

Statistical significance or p-value of t-test for testing mean difference has been widely used for the comparison of two groups. However, because of many issues that the statistical significance has, it has been intensively criticized in medical and social sciences. Consequently, effect sizes such as Cohen's d have been proposed as an alternative to statistical significance. Recently, strictly standardized mean difference (SSMD) has been proposed for the comparison of two groups with applications in data analysis in high-throughput screening experiments. In this article, from a theoretical basis, I explore the relationships between SSMD, standardized mean difference, and p-value of classical t-test for comparing two groups. The relationships among these measures hint that SSMD may serve as an alternative to statistical significance in medical sciences and as an alternative to traditional effect sizes in social sciences.