“It can be proven that most claimed research findings are false.”

A provocative essay from PLoS Medicine claims bias is present in most published research findings.

The study didn’t even need to include the Sokal Affair or random papers in its analysis, but it does discuss the so-called “file-drawer effect” (although some journals for negative results do exist).
There are also implications for the peer review process. An interesting read for those interested in the structure of scientific method.