1

I have different transcript expression data (continuous variable) and genotype data (categorical variable with 3 levels).

Most of the transcript data is normally distributed so I would like to use a parametric test to look for a significant difference based on the genotype.

However, one of the transcripts data is non-normally distributed and so I would have to use a non-parametric test to look for a significant difference.

I feel like if I was to make fair comparisons I would then have to do a non-parametric test on all of my transcript data rather than using two different types of tests. Is this correct?

Connor
  • 11
  • 2
  • I guess you could perform both kinds of tests, and check that their results are consistent. You can also have a look at this question : https://stats.stackexchange.com/questions/163915/why-would-parametric-statistics-ever-be-preferred-over-nonparametric – Camille Gontier Nov 24 '20 at 09:53

0 Answers0