P values for large sample size comparisons

Question

[Joshua]

Hi folks,

I am working with ~2500 genomes and am doing association analyses between groups of these genomes. I've been reading that large sample size can increase the likelihood of finding significant relationships and will decrease p values. (https://www.nature.com/articles/s41598-021-00199-5). Has anyone run across any methods used in the life sciences to counter this bias or have any advice for a stats novice?

Thanks, Josh

Answer 1

[swbarnes2]

Sounds like you need multiple testing correction.

See:

https://xkcd.com/882/