Hi everyone. I would like to run a Spearman's correlation test between two non-normal variables positively checked for monotone regression. I would like to be sure that some confounding factors are not influencing the result. Is that a way to check the presence of confounding factors and weigh their impact on the result?
Thank you in advance for your collaboration.
Dear Bradicich
Read "Handbook of Biological Statistics" John H. McDonald
Regards
Dear Bradicich,
It is difficult to give you the direct answer of your questions. You can control the confounding variables in various ways including Randomization, Restriction and Matching. It's totally depend on your experimental design and before the data gathering, or through statistical analysis after the data gathering process.
You can follow previous answers to know more about it. Matteo Bradicich
As Kolias says run partial correlations. Remember that Spearman equals to Person with ranks. So run partial correlation on ranks rather than on original values
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis