#FOAMPubMed 4: p values

In the previous blog we looked at how Type I Error is the false rejection of a null hypothesis.

THE MAXIMUM CHANCE WE WANT OF FALSELY REJECTING OUR NULL HYPOTHESIS IS 5%

This is a gold standard.

WE THEREFORE DESIGN STUDIES TO HAVE A LESS THAN 5% CHANCE OF FALSELY REJECTING OUR NULL HYPOTHESIS

A p value is a decimal showing the probability of falsely rejecting the null hypothesis. It will usually be given in a paper along with the results.

As we want a chance of less than 5% of falsely rejecting our null hypothesis the p value we want is p<0.05

Some studies want an even smaller chance of Type I Error and so design their study for p=0.01 (1% chance of falsely rejecting the null hypothesis) for example.

The p value we want will help shape our study, including sample size.

With p<0.05 we will have significant results - more of that in the next blog