'Conservative' p-values adjusted

[Schwann]

TL;DR

How to adjust p-values in such a way that they become more 'liberal'?

Hello everyone!

Could anybody recommend some strategy of p-values adjustment, as the distribution of my p-values indicates the presence of a big number of false negatives? Usually p-values are adjusted in order to overcome Type 1 errors (e. g. FDR or FWER estimation), but what I need to do is to control for Type 2 errors.

Any thoughts would be appreciated!

 

[Dale]

You need to do a power calculation. That will give you the information you need. However, be aware that doing so post-hoc is a statistical mistake. This analysis should be done before collecting data.