publication bias
Positive results preference
Publication bias is the tendency for scientific journals and researchers to publish studies with positive or statistically significant results while leaving negative or inconclusive findings unpublished.
This distorts the available evidence on any topic, making treatments or effects appear more reliable than they actually are.
The drug seemed effective until researchers accounted for publication bias and found the unpublished trials showed no benefit.
- findingsConclusions from investigation
- cherry-pickingChoosing only supporting examples
- selective reportingSharing only convenient findings
- journalsPeriodicals or personal logs
- articlesWritten works or items
positive results · methodology · reproducibility · transparency · registration requirements · integrity · meta-analysis · heterogeneity · winner's curse · suppressed · statistical significance · data analysis · data manipulation · peer review · underpowered · rejection · editorial boards · distortion
- biaseshalo effect, horn effect
- research ethicsresearch ethics, bioethics, ethics board, ethics training, fabrication
- Science Debatesclimate change, global warming, evolution debate, vaccine safety, creationism
- meta-analysismeta-analysis, heterogeneity, pooling, weighted average, weighting
- research biascherry-picking, cherry picking, selective reporting, suppression
- research validityvalidity, reliability, replicability, reproducibility, generalizability
bias, publish, publication, publishing, published, publisher, publications, unpublished, publishers