This function computes Cohen's d for a 2-group design (which is a standardized mean difference with a weighted variance standardizer) using a pooled-variance independent-samples t statistic and the two sample sizes. This function also computes the standard error for Cohen's d. The Cohen's d estimate and standard error assume equality of population variances.

stdmean2.from.t(t, n1, n2)

Arguments

t

pooled-variance t statistic

n1

sample size for group 1

n2

sample size for group 2

Value

Returns Cohen's d and its equal-variance standard error

Examples

stdmean2.from.t(3.27, 25, 25)
#>              Estimate        SE
#> Cohen's d:  0.9248957 0.2981791

# Should return:
#             Estimate       SE
# Cohen's d  0.9439677 0.298801