Search across guide titles, topics, keywords, and content throughout the Researcher's Field Guide.
Showing 226–250 of 482 results
Retraction does not necessarily mean every sentence in a paper is false. But identifying apparently unaffected information is not the same as restoring a retracted paper to ordinary evidential use.
Read Guide →Some corrections fix minor errors. Others change the result you were relying on. Learn how to reassess the study and update your own claims when the corrected evidence is materially different.
Read Guide →Research studies do not always reach the same conclusion, and disagreement does not automatically mean that one study is wrong. Learn how to compare apparently conflicting findings and judge what the wider body of evidence actually supports.
Read Guide →Methodological quality should strongly influence how much weight you give a paper because flaws in conduct can systematically distort its results. But quality should be assessed through relevant sources of bias, not reduced to a vague impression or simple numerical score.
Read Guide →A systematic review does not automatically outweigh a recent high-quality primary study. Its evidential value depends on the quality and currency of the review, the studies it includes, and whether the new study materially changes the evidence base.
Read Guide →Time-lag bias occurs when the speed of publication depends on what a study found. Positive findings may therefore enter the visible literature sooner, making early evidence look more favorable than the fuller evidence that emerges later.
Read Guide →Prestige can act as a shortcut for deciding which research deserves attention. When researchers notice or trust papers partly because of who wrote them or where they appeared, less prestigious but relevant evidence can become easier to overlook.
Read Guide →A note that preserves the source but loses the exact passage can still become difficult to verify. Capture page numbers, quotations, and other usable locators while reading so you can return to the evidence later.
Read Guide →Evidence should not influence a synthesis simply according to how many studies support each conclusion. Study limitations, relevance, precision, consistency, and the question being answered all affect how much confidence different findings warrant.
Read Guide →A literature review should distinguish what the evidence suggests from what it firmly establishes. Certainty depends on more than how many studies report similar findings.
Read Guide →Scientific consensus is the prevailing evidence-based judgment of a relevant scientific community, not a vote requiring unanimous agreement. Understanding what supports that consensus matters more than simply counting how many scientists agree.
Read Guide →Scientific consensus can be well justified while still leaving room for legitimate uncertainty. The challenge is recognizing when a dominant position appropriately reflects the evidence and when it prematurely marginalizes credible alternatives.
Read Guide →Sometimes a claim is repeated throughout the literature even though the source commonly cited for it does not actually provide the claimed evidence. The next step is not simply to reject the claim, but to determine what the source supports, whether other direct evidence exists, and where the mismatch arose.
Read Guide →A weak study should not become stronger simply because it supports your argument. Whether it belongs in a review depends on the review's eligibility criteria, while how much it should influence the conclusion depends on what its weaknesses mean for the evidence.
Read Guide →Before finalizing a literature review, imagine a knowledgeable reader who does not share your conclusion. What important evidence, qualification, or alternative explanation would they immediately notice is missing?
Read Guide →A field can produce many studies while repeatedly comparing the same options. Learn how to identify unanswered questions hidden behind familiar comparison groups.
Read Guide →You usually cannot prove that you found every relevant paper. What you can do is design, test, document, and cross-check your search well enough to make missing important literature reasonably unlikely.
Read Guide →Your final search should be recent enough that important new evidence is unlikely to make the submitted or published work materially outdated. Formal review standards may specify a maximum interval, while other projects require a judgment based on how quickly the field changes.
Read Guide →Finishing the writing does not freeze the evidence base. When new research appears, assess whether it changes your claims, synthesis, research gap, or conclusions, then revise proportionately rather than automatically rewriting everything.
Read Guide →A newly available database can improve an updated review, but it should not be added automatically. Assess whether it contributes relevant coverage, whether it overlaps existing sources, and whether it needs to be searched retrospectively.
Read Guide →A weak original search should not be reproduced merely for consistency. Identify the weakness, redesign the affected parts of the search, test the revised strategy, and determine whether earlier evidence must be searched again.
Read Guide →Not every literature review requires a formal protocol, but structured evidence syntheses generally benefit from prospective methodological planning. For systematic and scoping reviews, a protocol helps establish the question and methods before the emerging evidence can influence important decisions.
Read Guide →Preregistration creates a time-stamped public record of a planned review before its findings are known. It can strengthen transparency and help distinguish prespecified methods from later changes, but registration options and requirements differ between systematic and scoping reviews.
Read Guide →Two reviewers do not need to perform every systematic-review task independently, but independent duplicate assessment is an important safeguard at consequential stages. In major guidance such as Cochrane, final study inclusion decisions and outcome-data extraction require at least two people working independently.
Read Guide →There is no fixed number of years after which another review becomes justified. A recent review may still need updating or replacing if its question, methods, coverage, or evidence base no longer supports the answer researchers need.
Read Guide →