p-value cutoff for Wilcoxon rank sum test

[anagari]

Hi all,

I am using Wilcoxon rank sum test with continuity correction in order to determine if the differences in boxplot are significant.

Does anyone know about the p-value cutoff for a difference to be called as significant? Should I use default cutoff of < 0.05? p-value for most significant difference is < 2.2e-16.

If I use 0.05 cutoff, the differences that are not so significant by eye also pass through the threshold.

Can anyone suggest a p-value cutoff for wilcoxon test?


Thanks!