Peer review is the process of making sure we can trust the knowledge we publish.
In the peer review process, new research is sent to experts for them to check that the claims it makes are justified and strong enough to become public knowledge.
To get their work published, researchers sent it to research journals like Nature and Science that collect and share research from people all over the world. The research journals then organise the peer review process to help them decide whether or not this research will be published.
The reason we bother peer reviewing papers is because it is important that the common set of facts we all believe can be trusted. Peer review limits the reach and impact of poor research before it goes public, and makes sure that we don’t accidentally publish things that aren’t (to the best of our knowledge) true.
So how does it work?
A manuscript which explains what a group of researchers did, why they did it, and what they discovered, is sent to an expert who has no conflicts of interest with anyone who conducted the research. This expert then goes through the manuscript and sends the journal a report of their thoughts on what was good, bad, or needed improvement in the manuscript. The journal uses this report to decide whether the research should be published, rejected, or given back to the authors to make small changes before they make a decision.
Every journal has a slightly different way of running the peer review process. Some journals insist that both researchers and reviewers remain anonymous to one another until after the review is complete, others don’t. Some journals pay experts to review their manuscripts, some don’t. But the core method of the peer review process doesn’t change.
What’s the catch?
Unfortunately, peer review does not eliminate the possibility that research is wrong. Plenty of past research has been proven wrong, and current facts will probably be proven wrong sometime in the future, too. The peer review process does, however, prevent researchers from making claims that aren’t backed up by evidence or supported by our current knowledge.
The review process has practical limitations; it would be impossible to ask every expert in a field to review every new paper that comes out! So, instead of trying to find out whether all experts in a field agree, the peer review process normally asks around 2-4 experts to give their opinion on a piece of research. It might seem like these experts have a lot of power, but in the end it is the journal who decides what will be published, and not the experts.
This article is based on my experience with helping organise the peer review process at Diabetologia, where I currently work as Senior Editorial Assistant.
This article © 2025 by Ben Butterworth is licensed under CC BY 4.0