How was p value generated

[benralexander]

Mike says:

As another example, for a 4/0 (case/control) variant in IGF2BP2 I see a pvalue of .0262. This looks legit, but may or may not be since (absent a covariate effect) the pvalue is ~.125. Which points to the need also to specify somewhere clickable details like covariates in the model, whether this was joint or meta-analysis,