Half of social-science studies fail replication test in years-long project

Results from massive, ‘eagerly awaited’ initiative reinforce concerns about the credibility of science — but raise hope for solutions.

Email Bluesky Facebook LinkedIn Reddit Whatsapp X Access through your institution Buy or subscribe A massive seven-year project exploring 3,900 social-science papers has ended with a disturbing finding: researchers could replicate the results of only half of the studies that they tested 1 . Access options Access through your institution Access Nature and 54 other Nature Portfolio journals Get Nature+, our best-value online-access subscription $32.99 / 30 days cancel any time Learn more Subscribe to this journal Receive 52 print issues and online access $199.00 per year only $3.83 per issue Lea

related reading

• What's Wrong with Social Science and How to Fix It: Reflections After Reading 2578 Papersfantasticanachronism.com
• Reproducibility and transparency: what’s going on and how can we helpnature.com
• Psychology Research Is Mostly Fineastralcodexten.com
• It’s the end of science as we know it, and I feel finestatnews.com
• How Life Sciences Actually Work: Findings of a Year-Long Investigation - Alexey Guzeyguzey.com
• A Vision of Metasciencescienceplusplus.org
• How to Improve Both Scientific Innovation and Reproducibility at Once – Good Science Projectgoodscienceproject.org
• A Vision of Metasciencescienceplusplus.org
• Beware The Man Of One Studyslatestarcodex.com
• The rise and fall of peer review - by Adam Mastroianniexperimentalhistory.substack.com
• Age of Invention: Does History have a Replication Crisis?ageofinvention.xyz
• Scientific Fraud Is Slippery to Catch—but Easier to Combat | WIREDwired.com