Inclusion and exclusion criteria should be applied consistently, but consistency does not require mechanical interpretation. Learn how to distinguish justified judgment from making exceptions for studies you happen to prefer.
Read Guide →
Inclusion criteria can sometimes be changed after screening begins, but not simply because the studies you found are inconvenient. Any amendment should remain faithful to the review question, avoid being driven by study findings, and be documented transparently.
Read Guide →
Finding an influential or highly relevant-looking study outside your eligibility criteria creates a difficult choice. Its importance may justify discussing it, but not automatically including it in the evidence being synthesized.
Read Guide →
A study should not usually be excluded merely because it is old. A publication-date restriction is most defensible when time itself has a meaningful relationship to the review question, evidence, technology, policy, or context.
Read Guide →
A study should not be excluded merely because it comes from another country or a somewhat different population. The relevant question is whether those differences matter enough to the review question to affect eligibility or how the evidence should be interpreted.
Read Guide →
One study can produce several papers, and those papers should not automatically be counted as separate studies. Learn how to link related reports, combine their information, and prevent the same participants or data from influencing your review more than once.
Read Guide →
Duplicate publications can make one study appear to be several independent pieces of evidence. If reviewers fail to recognize the overlap, the same participants and findings may be counted repeatedly and distort the review.
Read Guide →
Two reviewers are often used to reduce errors and individual judgment in study selection, but the requirement depends on the review methodology and screening stage. Learn when independent duplicate screening is expected and what “two reviewers” actually means.
Read Guide →
A PRISMA flow diagram is designed to show how records identified through a systematic search moved through screening, retrieval, eligibility assessment, and final inclusion. It is appropriate for systematic reviews and many structured evidence syntheses, but not automatically for every literature review.
Read Guide →
When an abstract and full paper appear to describe different studies, do not choose one version automatically. Verify that they are the same report, identify the discrepancy, and base eligibility on the most complete reliable information available.
Read Guide →
Two papers are not necessarily two independent studies. Learn how to compare study identifiers, participants, recruitment periods, settings, methods, and other clues to determine whether publications report the same underlying study.
Read Guide →
One study can generate several useful papers, but publication count should not be mistaken for study count. Learn how to preserve information across reports while keeping the underlying study as the unit of evidence.
Read Guide →
When one study has several publications, the primary report is usually the report chosen as the main source for study results, not simply the first or most cited paper. Learn how to make and document that choice.
Read Guide →
Companion papers can contain valuable outcomes, methods, harms, and follow-up data that the main publication omits. The key is to use those reports as sources for one study without treating their shared participants as independent evidence.
Read Guide →
Different papers from the same study may legitimately report different outcomes, but those outcomes still come from one underlying investigation. Learn how to map, select, reconcile, and synthesize them without losing information or creating duplicate evidence.
Read Guide →
A preprint and a later journal article may be separate reports of the same underlying study. Learn how to link versions, inspect meaningful changes, and avoid treating one study as two independent pieces of evidence.
Read Guide →
You rarely need to read every research paper from the first word to the last. Learn a practical multi-pass method for deciding what deserves attention, extracting the important evidence, and knowing when a paper requires a deeper read.
Read Guide →
You rarely need to read every research paper from the first word to the last. The depth and order of your reading should depend on why you opened the paper and what you need to learn from it.
Read Guide →
Start by orienting yourself to what the paper studies and whether it is relevant to your purpose. The title and abstract are usually useful first stops, but what you read next should depend on what you need from the paper.
Read Guide →
There is no single best section to start with in every research paper. The abstract is useful for orientation, but your next move should depend on whether you need context, methodological detail, evidence, or interpretation.
Read Guide →
Not every paper you find deserves the same amount of attention. Screen its relevance first, then consider whether the evidence, methods, and role it may play in your research justify a deeper reading.
Read Guide →
You do not need the same depth of understanding for every research paper you encounter. The more consequential the paper is to your argument, methodology, or interpretation, the more deeply you need to understand the parts on which you rely.
Read Guide →
The discussion deserves serious attention, but it is the authors' interpretation of the evidence rather than the evidence itself. Compare its major claims with the results, methods, limitations, and plausible alternative explanations before accepting them.
Read Guide →
A paper outside your discipline can be difficult even when you are an experienced researcher. Build enough disciplinary context to understand the question, terminology, methods, evidence, and assumptions that matter to why you are reading it.
Read Guide →
You do not need to master every unfamiliar method before you can learn from a paper. First identify exactly what you do not understand, determine whether it matters to your purpose, and investigate it deeply enough to interpret the evidence responsibly.
Read Guide →