Linked from

The 45 pages that link to Publication bias, each with the reason it gives.

• Evidence-based medicineRelated: It can make the published evidence base overstate benefits or understate harms.

• Scholarly peer reviewRelated: Editorial and reviewer preferences can contribute to which results enter the published record.

• Replication crisisCompared with: It is one proposed cause rather than the crisis itself.

• ReproducibilityRelated: Selective publication can hide failed repetitions and exaggerate confidence in findings.

• EvidenceRelated: It can distort the available evidence by hiding null or unfavorable findings.

• Systematic reviewRelated: Unpublished or hard-to-find findings can make a review’s evidence base unrepresentative.

• P-valueRelated: Preference for significant results distorts the evidence represented in published p-values.

• Sampling biasCompared with: It filters which studies become visible rather than which individuals enter a study.

• Meta-analysisRelated: Missing studies can make a pooled effect appear stronger or more certain.

• Selection biasBroader topic: It selects studies into the evidence base rather than participants into an individual study.

• Multiple comparisons problemRelated: Selective publication can amplify the apparent success of chance findings.

• Stereotype threatRelated: It can distort the published evidence used to estimate the effect's typical strength.

• Null hypothesisRelated: Null findings may be underreported, distorting the evidence available across studies.

• Risk compensationRelated: Selective reporting can distort estimates of how often compensation occurs.

• Effect sizeRelated: Selective publication can make reported effects appear larger than they are.

• ReplicationRelated: When failed replications remain unpublished, the published record overstates how often findings recur.

• Insect declineCompared with: A literature dominated by alarming findings may misrepresent the balance of measured trends.

• StatisticsRelated: Selective reporting can skew the evidence available for statistical synthesis.

• ParapsychologyRelated: It can make a literature of small positive findings overstate evidence for psychic effects.

• Animal modelRelated: Unpublished negative animal studies can make preclinical evidence appear stronger than it is.

• Ecological validityRelated: A distorted published record can undermine judgments about whether findings hold in practice.

• Placebo-controlled trialRelated: Unpublished placebo-controlled trials can distort estimates of treatment benefit and harm.

• Scientific reproducibilityRelated: Selective publication hides failed repetitions and distorts the evidence base.

• ExperimentRelated: Selective reporting can make experimental effects seem more dependable than they are.

• Drug repurposingRelated: Unreported negative repurposing studies can make candidate drugs appear more promising than the evidence warrants.

• Quantitative researchRelated: Selective reporting can make the published quantitative evidence unrepresentative.

• Scientific misconductRelated: It can distort the literature even when no individual commits misconduct.

• Reproducible researchRelated: It can hide unsuccessful attempts that matter when assessing a finding's reliability.

• Scientific publishingRelated: Selective publication can make the published literature misrepresent the full body of research.

• Type I and type II errorsRelated: Selective attention to significant results can conceal false positives and studies with type II errors.

• BiostatisticsRelated: Selective publication can skew pooled evidence and clinical conclusions.

• Clinical researchRelated: Unpublished or selectively reported findings can distort the apparent evidence base.

• Ganzfeld experimentRelated: A literature containing disproportionately positive trials can exaggerate apparent above-chance performance.

• Prediction intervalRelated: Selective evidence can make estimated heterogeneity and predictive ranges misleading.

• Case reportRelated: Unusual or successful cases are more likely to appear, distorting the published picture.

• Null hypothesis significance testingRelated: Preference for significant findings can make the published evidence unrepresentative.

• Esther DufloRelated: Selective publication can distort the evidence base used to judge interventions.

• PLOS ONERelated: A technical-validity threshold can make room for sound findings with limited perceived impact.

• Open-label placeboRelated: Small or positive open-label placebo studies may overstate effects if null findings remain unpublished.

• Statistical hypothesis testRelated: Selective publication of significant tests can exaggerate apparent evidence.

• Psychological researchRelated: It can make the published psychological literature misrepresent the full evidence.

• ResearchRelated: It can distort the visible record of findings.

• Birthday-number effectRelated: Selective reporting can make small or inconsistent preference effects appear more dependable.

• Funding of scienceRelated: Selective funding and publication incentives can favor positive or striking findings.

• Literature (scholarly publications)Compared with: The published record may not represent all studies conducted on a topic.