Reliability of Difference Between Uncorrelated Means (Nᡖ)
One vs Two tailed tests
- In the case of a 1 tailed test, we look up the tabled value for .05 rather than .025.
- When using a 1 tailed test we predict the direction of difference
- A smaller difference is neeeded for the 1 tailed test.
Advantages & Disadvantages
- A smaller difference is needed when using a 1 tailed test.
- To use a 1 tailed test we must be able to predict the direction of difference between the groups
- If we predict, we must ignore difference in oposite direction if we get it.