t-test for independent samples A two-tailed t-test for independent samples (equal variances assumed) showed that the difference between yes and no regarding the dependent variable ABUSE was not statistically significant, t (5622) = 0.51, p = .613, 95%, p < .05 confidence interval [-0.46, 0.78]. Thus, the null hypothesis was not rejected. The effect size d was 0.03 (equal variances assumed). With d = 0.03, there was a very small effect