When a researcher has two samples with two means, one of the assumptions for two samples is independence of between the two samples. What does this