Samples are to be taken from two independent normal populations, one with variance σ2 = 10, the other with a variance twice the magnitude of the ?rst one. If the difference between the two population means is to be estimated to within ±2 with a two-sample test at the α = 0.05 signi?cance level, determine the sample size required for the test to have power of 0.9 or better, assuming that n1 = n2 = n . State any other assumptions needed to answer this question.