For testing the difference in means of two independent samples with n1 and n2 items respectively, the t-statistic is computed with which degrees of freedom?

For an independent-samples t-test comparing two means, the degrees of freedom are n1 plus n2 minus 2.

One degree of freedom is lost for each of the two sample means that are estimated.

n1 plus n2 minus 1 applies to a single-sample situation, not two independent groups.

n1 plus n2 ignores the estimation of the means and overstates the degrees of freedom.

n1 plus n2 plus 1 has no basis in the pooled-variance t-test formula.

The pooled t-test assumes equal variances across the two populations.