By Jayson Chaw. Tue Jul 28.
What actually happens between a scientist finishing a study and that study appearing in a journal, and why the process has real limitations.
When a news article mentions "a peer-reviewed study", it is usually offered as a mark of credibility, and there is good reason for that. But peer review is often treated as a vague stamp of approval rather than understood as a specific process with clear steps and real limitations. Knowing how it actually works makes it much easier to judge how much weight a particular study deserves.
The basic process
Peer review is the process by which other experts in the same field assess a study before it is published, checking whether the methods are sound, the analysis is appropriate, and the conclusions are reasonably supported by the data presented.
A researcher submits their study to a journal
An editor checks whether it fits the journal's scope and meets basic quality standards
The editor sends the study to independent experts in the relevant field, the "peers"
These reviewers examine the methodology, the data, and the conclusions, and write a report
The editor decides to accept, reject, or request revisions based on the reviewers' feedback
If revisions are requested, the researcher responds and the study may be reviewed again before publication
What reviewers are actually checking
Reviewers are not simply checking whether they agree with the conclusion. Their job is closer to that of a demanding examiner, asking whether the method was appropriate for the question being asked, whether the sample size was large enough to support the conclusions drawn, whether the statistical analysis was applied correctly, and whether the authors have acknowledged the limitations of their own work rather than overstating what the results show.
Why this slows science down, on purpose
Peer review deliberately introduces friction into the publishing process. That friction is the point: it gives independent experts a chance to catch flawed methodology, misinterpreted data, or unsupported claims before they reach a wider audience, rather than after the damage of a misleading headline has already been done.
What peer review does not guarantee
It is important not to overstate what peer review achieves. A peer-reviewed study is not automatically true, and treating it that way misunderstands the process.
Reviewers check methodology, but they cannot detect fraud or fabricated data if it is presented convincingly
A single study, even a well-conducted one, can still produce a result that later fails to be repeated by other researchers
Reviewers can disagree, and the same stu…