My once great enthusiasm with modern science has diminished ever since noticing a predictable pattern in almost all biological research. Substance X shows promising preliminary findings, a multitude of further studies support substance X, some studies don't support it. A meta-analysis then examines the materials & methods and concludes most trials were poorly conducted. Further investigation concludes research into substance X was funded indirectly via companies (which will often not appear in Conflicts of Interest sections), and 10 years later, Substance X has lost the hype in place of Substance Y, which will repeat the process.
It got me questioning, with so many meta-analyses and review studies demonstrating that the majority of published journal articles in reputable journals are not replicable, and the findings made are often incorrect, are there any studies to see if scientists can even get the basic methodology itself right?
What I mean is, a triple blind, cross-over, placebo controlled study, but this time instead of a new drug to compare the placebo pill to, we have the identical placebo pill itself in a different colour. The researchers are told that it's an active ingredient when it's not. This study will tell us which scientists can even get the basic methodology right before we should take their future research seriously.
Do any such studies exist? If not, why not? This surely seems to be a pertinent issue given the poor methodology and lack of replication that is apparent in most research.