What is it about?
I show that the low replication probability of P-values is shared by other probability statements.
Featured Image
Why is it important?
If a P-value carried with it a high probability that it would be replicated, then this would be a form of inferential anticipation that would be unfortunate. The fact that a P-value of 0.048 (say) will not be replicated with high probability is not an undesirable property but a desirable one shared with other inferential devices.
Read the Original
This page is a summary of: A comment on replication, p‐values and evidence S.N.Goodman, Statistics in Medicine 1992; 11:875‐879, Statistics in Medicine, July 2002, Wiley,
DOI: 10.1002/sim.1072.
You can read the full text:
Resources
Contributors
The following have contributed to this page