Suppose I have a 2 by 2 table and I am trying to conduct a Fisher's Exact Test. I am wondering when the two-sided test is used versus a one-sided test. In programs like R
, they have greater
, less
, and two-sided
. In most cases, I would only use either greater or less depending on the direction. However, when is the two-sided version ever used? Thanks.
Solved – When is a one-sided test used versus a two sided test in a Fisher’s Exact Test
contingency tablesfishers-exact-testhypothesis testing
Related Question
- Solved – the difference between two-sided and one-sided fisher exact test
- Solved – Computing the power of Fisher’s exact test in R
- Solved – Fisher’s exact test in RNA-Seq
- Solved – Fisher’s exact test – help understanding the p-value
- Solved – 2-Sample Proportions z-test vs Fisher’s Exact Test
- Solved – Two-sided vs One-sided p-value: is there a special case with Fisher’s Exact test
Best Answer
If you have a very strong assumption before exploring your data that the values of one group will be either lower or equal to the other group (
alternative="less"
) or that the values of one group will be either higher or equal to the other group (alternative="greater"
) you can use a one-sided test to increase the power of this test:A two-sided test:
A one-tailed test:
https://stats.idre.ucla.edu/other/mult-pkg/faq/general/faq-what-are-the-differences-between-one-tailed-and-two-tailed-tests/