In: Statistics and Probability
When Comparing means using the 2 Sample t Test:
(A) How many tails are appropriate for this test and why?
(B) What assumptions are made by your test?
(C) In general, what form of data model is this analysis most appropriate for? (How many IVs and DVs?)
We can use two tailed test for this situation . Here in null hypothesis we will state both the means are same and for alternative hypothesis will state both the means are different . The value of t critical varies for one tailed or two tailed test .
We are using two tailed test here because by that we get bigger confidence interval and hence the probability of getting the t score within the interval increases .
The populations from which the samples have been drawn should be normal