I am working on some physiological traits of plant to which its expensive to raise the sample size beyond n=3. so kindly suggest some test which is applicable to test such group(2)
Go through this article, may be it's useful to you.
doi: https://doi.org/10.1136/bmj.a3166
Thank you Busari Yusuf for you advice but i have one more doubt that whether i should have to use any post hoc test after t test or not.
Post hoc test is done to unravel many statistical problems o. And it is used to compare 3 sampled groups or nore. You are
Hello Manish,
Unless the magnitude of difference between groups/conditions is substantial, any statistical test you might select will have very poor statistical power to determine that the difference is different from zero. That's unfortunately the case with very small N studies.
As you have only two groups, there would be no need for a post hoc test beyond the group comparison test itself.
Good luck with your work.
There are various "exact tests" that might be appropriate to your situation. A specialized software package called StatXact could be helpful. Some of these tests are also available through general purpose statistical software, such as SAS, SPSS, and R. These tests avoid parametric assumptions that are usually not tenable with very small samples. But they are not a magic solution to the need for substantial effect sizes, as rightly pointed out by David Morse .
Only nonparametric methods are helpful.
Also Single case follow-up studies ABAB design can be adopted. A minimum of at least 5 or more is required for nonparametric methods
In addition to traditional t-test that is theoretically not restricted to small samples, you may calculate the relative mean effect size (RMES) that may give you an idea about the "real" difference so may help make a scientific inference. As for RMES, please refer to a preprint, "Exceedance probability analysis: a practical and effective alternative to t-tests" on ResearchGate:https://www.researchgate.net/publication/348692325_Exceedance_probability_analysis_a_practical_and_effective_alternative_to_t-tests.
hi,
The answer to your question depends on what criteria you want to compare between the two groups
for example the t-test is for compare the mean of two groups
for another example if you want to compare the variance of two groups you can use the ratio variance test
if you want to do these test with software i suggest you to use R program.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size