The director of the National Institutes of Health, Dr. Francis Collins, is worried about the lack of reproducibility and 'secret sauce' in a large number of studies funded by their $30 billion government agency.
Fraud happens everywhere, as does cherry-picking of results, but the more scientific the field, the less it happens. It's hard to get 2,000 people to be fraudulent about an experimental physics result while a lone psychologist writing about surveys of college students is difficult to catch.
Carnegie-Mellon and Stanford scholars say they have a solution for Dr. Collins, at least - use