Large replication projects answer the wrong question, because most individual studies do not matter either way.

Added to korrents 7 Sept 2026 · How quotes work · Something wrong? Tell us

Shared subject first, then similar wording. They may agree or disagree.

• A tiny, unreplicated study should not be publicised to millions of people as though it were settled. SRStuart Ritchie11 Oct 2022
• The answer to weak studies of programming practice is to fund better ones, not to conclude that the question cannot be measured. DLDan Luu7 Nov 2014
• Only 30 to 50 percent of product experiments win, because consumer behaviour is too unpredictable to be predicted from a hypothesis. ACAlbert Cheng5 Oct 2025
• A range of studies from many places and methods showing the same result does not necessarily make a strong signal. GTGary Taubes30 Nov 2025
• The replication crisis is not an aberration; many of the things we believe are wrong SRStuart Ritchie27 Dec 2022

See this claim on the map