Psychology › Schools & history › The replication crisis
Reproducibility Project: under 40% replicated
2015· Finding or experiment
Well replicatedStrong, repeated evidence.
A 2015 effort in which 270 researchers repeated 100 published psychology studies; only about 36% produced significant results again.
See it on the Psychology map →What it says
Led by Brian Nosek's Center for Open Science, teams redid 100 experiments from 2008 issues of three leading journals, using original materials where possible and consulting the original authors. Of the originals, 97% had reported significant results; 36% of the repeats did. Average effect sizes fell by about half. Cognitive psychology fared better (about half replicated) than social psychology (about a quarter). Published in Science, it became the crisis's headline number.
Why it matters
It turned scattered failures into a systematic estimate, pushed journals and funders toward reform, and inspired similar projects in economics and cancer biology.
The evidence
Status: well replicated as a measurement. All protocols, data and analysis code are public, so anyone can check them; later large projects found broadly similar rates.
Open questions
What counts as 'replicated': different criteria give rates from about 36% to 47%, and no single cutoff is agreed.
Connections (2)
▸Reproducibility Project: under 40% replicated measured the scale of The replication crisisWell replicated
The connection
Before 2015 the crisis was a string of individual failures and arguments. By repeating 100 studies chosen systematically, the project gave it a number: about 36% replicated, with effects about half as large. That figure made headlines worldwide and became the standard shorthand for the problem.
Who showed it and when
Coordinated by Brian Nosek's Center for Open Science in Charlottesville, Virginia; 270 coauthors; published in Science in August 2015.
The evidence
Every protocol, dataset and analysis is public on the Open Science Framework, and original authors reviewed most replication designs in advance.
Open questions
Daniel Gilbert and colleagues argued in 2016 that design differences and chance made the rate look worse than it was; the project's authors disputed this.
▸Reproducibility Project: under 40% replicated found weakest replication in Social cognition: how we judge peopleWell replicated
The connection
The 2015 Reproducibility Project re-ran 100 studies from leading journals. Social-psychology studies, many on social cognition topics such as priming and judgment, replicated only about 25 percent of the time, versus about 50 percent for cognitive studies, casting doubt on many celebrated findings about how we judge people.
Who showed it and when
Published in Science in August 2015 by the Open Science Collaboration, about 270 researchers led by Brian Nosek of the University of Virginia, re-testing studies published in 2008.
The evidence
Replications used original materials where possible and registered plans in advance; all data are public. Later multi-lab projects found similar patterns.
Open questions
Some argue social effects are more sensitive to time and place, so failures reflect changed contexts; others say many original studies were simply too small.