Peer reviewed study published in Science Journal turns out to be fraud.

[Jenny A]

Peer review is a process in which scholars in the same field review a manuscript before publication in a a peer reviewed journal.  The manuscript is reviewed to see that the methodology and procedures described in the manuscript are sound.  It is not generally speaking a fact checking activity.   Generally speaking it assumes the person submitting the manuscript is honest.  What's being reviewed is the quality of his methods and procedures and the validity of his conclusions based up those methods and procedures.

Independent replication does not generally occur until after a study has be peer reviewed and published.  It's hard to attempt to replicate what you don't know about.   It is at this point other scientists will attempt to replicate the study using the same and/or different methods.   The more controversial or unlikely a finding, the more likely it is that others will attempt to replicate it. But generally speaking the purpose of independent replication is not to uncover actual fraud, but rather to determine if honestly reached conclusions hold up when examined more closely.  {edit: or using a larger sample size} Often they don't though the mistakes are in the vast majority of cases honest ones rather than fraud.  For that reason a single study is rarely considered conclusive in the sciences.  (Though the popular press  both on the left and right does like to treat single studies as if they were should the story be good enough.)


Though the system is not designed to detect fraud, that a fraud will eventually be discovered is almost inevitable given the way the process works as fraudulent results simply are not replicable.

In this particular case the methods described in the paper were sound, and the paper was published. The paper got a lot of press because it was good news for a number of interests.  But it failed the test of independent replication so badly that fraud was suspected and indeed found to be the case.  There is nothing "chance" about it.  

So what did we learn?  We learned a scientist falsified data and that he was found out.  Notice that he was found out not by the press but by other scientists.