In: Statistics and Probability
When do you use a one-tailed test? When do you use a two-tailed test? How is SPSS used to perform the calculations needed for each?
We use the one-tailed test when we need to check or test the claim or hypothesis which is directional in nature. This means if the alternative hypothesis of the test includes ‘less than’ or ‘greater than’, then we use one-tailed tests. We use the two-tailed test when we need to check or test the claim or hypothesis which is non-directional in nature. That is, if the alternative hypothesis of the test includes ‘not equal to’, then we use two-tailed tests. When we perform one-tailed tests and two-tailed test by using SPSS, then we need to use some setting or calculations. For most of the tests, SPSS provides the results for the two-tailed test as a default. Most of the times, SPSS do not provide results for one-tailed tests and in such cases, we need to calculate the p-value from this output.
Examples of one tailed and two tailed test is given as below:
H0: µ = 10 v/s Ha: µ < 10 (One tailed /lower tailed /left tailed /directional test)
H0: µ = 10 v/s Ha: µ > 10 (One tailed /upper tailed /right tailed /directional test)
H0: µ = 10 v/s Ha: µ ≠ 10 (two tailed /non-directional test)