I have previously explained what the scientific method is. However, finding answers to questions is only part of the task of a scientist. The next step after finding answers is telling the world what you have found. This is not an easy task. For scientists, publishing peer reviewed articles is essential. But what is peer review?
There is a big difference between peer reviewed journals and other publication methods. The name “peer review” already tells part of the story. A “peer” is a colleague, and “review” is that it is being assessed. So in short it means that the quality and validity of your research is being assessed by colleagues. This is very different from for example this blog, on which I can post anything without someone performing a quality check (of course I do everything I can to assure high quality on my blog too). However, peer review is not the only part of publishing in peer reviewed journals. The first obstacle is the editor.
Calling the editor an obstacle sounds a bit cruel, but they are the first people you have to convince that your research is worth publishing. The editor works for a journal and assesses how suitable papers are for his journal. If I write a paper about the immune response in beetles and send it to the journal “Plant Science” then the editor will say that my beetle research in not suitable for this journal. However when I send the paper to “Developmental and Comparative Immunology” the editor will agree with me that it suits his audience.
A second aspect of editors is that they are responsible for keeping the quality of the journal high. This means that they will not accept any paper that fits in the description of the journal. If the research you present isn’t innovative or new enough than the editor will likely reject the paper. How often the editor will reject depends on the journal. The high impact journal “Nature” for example, publishes scientific results from any scientific field but only the most surprising / exciting / innovative findings will be accepted. They reject about 90-95% of all papers that are submitted to them. To the contrary, “Plos One” accepts all papers that are scientifically sound. No matter how unexciting they may be.
Once you have convinced the editor that he really wants your paper then the review process starts. The editor will find scientists in the same field as you are in (you may suggest suitable reviewers) and sends the paper to between 2 and 4 of your colleagues.
The reason for peer-review is simple. Scientific publications need to be sound. When your colleagues read the paper they are able to provide valuable feedback which in most cases improves the quality of the paper. They look at the data you gathered and which conclusions you draw from them. Then the reviewers will provide feedback on whether the experiments you have performed are appropriate and the conclusions you draw from them are backed up by the data. Often they will ask for more experiments to verify that what you say is true, or ask to rephrase certain parts of the paper. Although peer-review improves the quality of the paper, scientists (and me among them) aren’t always happy with the comments of their peers. They might ask impossible experiments, tell you that things are missing even when they are there and so on. It is important however that reviewers feel free to comment on the paper without being afraid that the author of the paper will get angry with them. So in general peer-review is performed blind, the author of the paper does not know the identity of the reviewer.
Once the reviewers have read the paper and commented on it, they will advise the editor on whether to publish it or not. The advice looks like this:
1) Accept without modification (very rare!)
2) Accept with minor revisions (change some text here and there)
3) Accept with major revisions (additional experiments necessary)
With the advice they of course add a document stating what changes are necessary or why they reject the paper. The editor makes a decision based on these reviews. If 2 reviewers for example give the advice: “Accept with minor revisions” and “Reject” he could decide to accept after revisions. On the other hand, if major revisions are required, then another round of review is common. After the author has performed the additional experiments, the editor will send it again to the same reviewers and ask whether they are satisfied with the new experiments. When rejected there is not much choice but to try again with some other journal. It would be wise to do something with the reviewers’ comments because they might also be selected by the other journal. In the end, after some hard work, the paper will be accepted for publication. How long it takes varies a lot and can take anywhere between 2 months and a year.
Although peer-review is essential to assure the quality of research, it is by no means a guarantee that once published in a peer-reviewed journal, research is always true. Some reviewers are more thorough than others. Furthermore, reviewers can only assess the data as it is presented to them and see whether it was analyzed the correct way. They cannot verify that the data are correct. That would require redoing the experiments themselves and nobody has time to do that. This means that it is very difficult to find out if a researcher made mistakes with processing the data or has purposefully left out data that didn’t fit the hypothesis. Luckily, most researchers are passionate about finding the true answer to their questions. Furthermore, sooner or later other researchers will follow up on your research and the mistakes will surface. So although in general scientific research is of very high quality: always stay critical!