Count replications and open data in hiring and promotion
Scientists are promoted for novel discoveries, not for checking others' work or sharing data. Changing what universities reward would change what scientists do.
Career incentives drive the reproducibility problem: novelty and journal prestige are rewarded, replication and data sharing are not. Institutions signing DORA have pledged to reduce reliance on journal metrics. A concrete step is a structured section in tenure and promotion dossiers for replication studies conducted, data and code shared, registered reports and independent verification of one's own findings, weighted explicitly in decisions.
- Preclinical results do not reproduce · Fewer than half of landmark cancer biology findings reproduce when someone else tries.
- Incentives reward me-too drugs and marginal gains · The system pays the same for a drug that adds two months as for a cure, so companies race to copy rather than to cure.
Pages like this
not linked directly; found by shared links- IdeaEvery cancer biology PhD begins with a funded replication of a published finding
Shares Funders set aside a fixed share of budget for independent replication, Preclinical results do not reproduce.
- IdeaBounties for documented failed replications of high-impact findings
Shares Funders set aside a fixed share of budget for independent replication, Preclinical results do not reproduce.
- IdeaIndependent replication of the key experiment before first-in-human academic trials
Shares Funders set aside a fixed share of budget for independent replication, Preclinical results do not reproduce.
- IdeaPromote academics for trials completed, data shared and findings replicated
Shares Preclinical results do not reproduce, Incentives reward me-too drugs and marginal gains.