THE REPLICABILITY PROBLEMS IN SCIENCE: IT’S NOT THE P-VALUES FAULT

YOAV BENJAMINI – TEL AVIV UNIVERSITY

Significance testing, and the p-value as its symbol, have become the statistical scapegoat for the replicability problems in science. Instead, I shall argue that the two main statistical obstacles to replicability are (i) unattended inference on the selected, and (ii) ignoring the relevant variability. I shall devote most of the talk to the first challenge: reviewing current and emerging approaches to selective inference, and presenting some recently developed methodologies, with examples from genomics, medical research and experimental Psychology.