Open Thread: Publication Bias Edition

Conference season is almost over for me; thanks for your patience. In the meantime, please enjoy this picture of a small mushroom in Saxony, and note also this sequel to the mask replication study I posted about on Sunday:

Jay Bhattacharya @DrJBhattacharya

Why did the @CDCMMWR reject a reanalysis by @TracyBethHoeg and @AChandra_TO of a mask study it previously published? The reanalysis used more data & same methodology but failed to replicate the original paper. Publication bias is not usually quite so blatant.

Kyle Lamb @kylamb8

One study found in the CDC MMWR has been pushed by mask propagandists to show mask mandates work in schools despite a cutoff of September 4…early in a school year. A new study took that methodology and extended endpoint a month and increased sample size 6x. Guess what happened? https://t.co/1RzmzBNYUZ

The CDC’s Morbidity and Mortality Weekly Report were totally happy publishing the original paper on Pediatric COVID-19 Cases in Counties With and Without School Mask Requirements, despite serious defects. Yet they refused to publish the much larger replication study, which failed to find any protective effect from masking when using the same methodology on a much larger dataset. Publication bias is a pervasive problem with all Corona research, and it is the major reason why most mask studies are garbage.