What it does check
Whether the question is worth asking, whether the method could plausibly answer it, whether the analysis matches the design, and whether the conclusions follow from the results as reported.
Those are substantial. A paper that has been through competent review has had somebody with subject knowledge look for the obvious problems, which is more than can be said for most of what circulates.
What it does not check
Whether the data are real. Reviewers see a manuscript, not a dataset, and almost never request raw data. Fabrication is generally caught after publication, by readers, which is why retractions exist at all.
Nor does it check whether the finding replicates, which by definition cannot be known at the time of review.
Review quality varies enormously
Two reviewers at a rigorous journal reading closely is a different process from two reviewers skimming for a publisher with a fast turnaround. The label is identical in both cases.
This is what makes peer-reviewed a weak signal on its own. Where a paper matters to your argument, the journal, the reviewers' evident engagement and the study's own transparency tell you more than the label does.
Predatory publishing exists
Some outlets charge a fee and publish essentially anything, while describing themselves as peer-reviewed. They frequently mimic legitimate journal names closely enough to be mistaken for them.
Signs worth noticing: an editorial board you cannot verify, a scope covering everything, an implausibly fast turnaround advertised as a feature, and unsolicited invitations to submit.
Post-publication is where a lot happens
Commentary, correspondence, failed replications and corrections all arrive after review, and they are frequently where a paper's real standing gets established.
Looking at what has cited a study since publication is often more informative than knowing it was reviewed, particularly for anything more than a few years old.
What to do with all this
Use peer review as one input rather than as a verdict. Where a claim matters, look at the method yourself, check the journal is what it appears to be, and see what the field has said since.
That is exactly the appraisal a graduate rubric is asking for, and it is why appraisal criteria exist alongside requirements about source type.
It is still a meaningful filter
None of this argues for ignoring it. Peer-reviewed work is on average considerably more reliable than unreviewed work, and a rubric asking for it is asking for something sensible.
The point is that it is a floor rather than a guarantee, and treating it as a guarantee is what leads to citing a reviewed paper that a moment's reading would have shown was unsuitable.
Say what you checked
Where you have looked beyond the label — noted a small sample, a short follow-up, a setting unlike yours — say so in the sentence. That is the visible difference between a paper that cites and one that appraises.
It also protects you. A limitation you named yourself cannot be used against you by somebody who noticed it after you did.
The habit underneath
Every issue of this letter reduces to the same instruction: look at the thing itself rather than at the label on it. Open the source, read the method, check the version, follow the citation back.
None of it is difficult and all of it is skipped under deadline pressure, which is precisely why doing it consistently is worth more than any individual technique.
Questions this raises.
Should I only cite peer-reviewed sources?
Follow your rubric, and understand what it is reaching for. The instruction usually means avoid unreviewed opinion rather than exclude everything else, since guidelines, statistics and standards are none of them peer-reviewed and are frequently the correct citation. Match the source type to the job it is doing.
How do I check a journal is legitimate?
Look at whether it is indexed in the major databases for your field, whether its editorial board members exist and acknowledge the role, and whether its published articles look like research. Your librarian can settle it in minutes, and that is a better route than any single online list.
What is a predatory journal?
An outlet that charges to publish and performs little or no meaningful review while claiming otherwise. The commercial model is the fee rather than the readership. They often copy the names of established journals closely, so a citation that looks familiar but slightly wrong is worth checking rather than assuming.
Does peer review catch fabricated data?
Rarely, because reviewers read a manuscript rather than a dataset and almost never see raw data. Fabrication is usually detected afterwards by readers who cannot replicate a result or who notice inconsistencies in the reported figures. This is precisely why retractions exist and why checking for them matters.
Is a preprint worthless then?
Not at all, and it should be labelled as what it is. A preprint is research that has not yet been reviewed, which makes it weaker evidence and sometimes the only evidence available on a recent question. Cite it explicitly as a preprint and check whether a reviewed version has since appeared.
Dr. Sigrid Nyholm
PhD, Nursing, on the evidence-based practice side of the hall. Writes for The Footnote and coaches the students who bring this work to the desk. The rest of the specialists.
Elsewhere in the hall.
Send the assignment and the deadline.
A specialist credentialed in your own field reads what you actually have and replies within two hours with a plan and a price. The reading costs nothing, and if the honest answer is that we are not the right hall for it, you will get that instead.
