Compute the standardized mean difference between two groups, either from descriptive statistics or from a reported t test. Includes the Hedges' g small-sample correction and conversion to r.
Uses d = t × √(1/n₁ + 1/n₂) for an independent samples t test.
Effect sizes computed automatically. ReliCheck Quanta reports d, g, eta squared, and omega squared alongside every relevant test, formatted for APA. Free to try on your Mac.
Get QuantaCohen's d expresses a mean difference in standard deviation units. A d of 0.50 means the group means sit half a standard deviation apart, regardless of the original measurement scale. That makes d comparable across studies and essential for meta-analysis and power analysis.
Cohen's rough benchmarks are 0.20 small, 0.50 medium, and 0.80 large. Field context matters more: in education research, an intervention effect of d = 0.20 on standardized achievement can be substantial. Use benchmarks as a starting point, not a verdict.
Deciding what counts as a meaningful effect in your field is a judgment call, not a calculation, and it is the part reviewers question most. Our statistical consulting includes reading an effect size against the literature you are writing into.
Quanta