---
title: "Uri Simonsohn"
type: "entity"
entity_kind: "person"
status: "hub"
canonical_name: "Uri Simonsohn"
aliases: []
first_seen: "2026-07-11T00:00:00.000Z"
writer_model: "claude-opus-4-8"
connects_to: ["data forensics","excessive similarity","Data Colada","research integrity","R. A. Fisher"]
seek_code_commit: "a619c8a"
---


Uri Simonsohn is a behavioral scientist (Esade Business School, Barcelona) and a co-founder — with Leif Nelson and Joseph Simmons — of Data Colada, the research-integrity blog that has driven several high-profile retractions in psychology. He is known for methods that catch questionable research from published statistics alone: "p-curve" analysis for detecting p-hacking, and a fabrication-detection technique that flags data whose summary statistics show "excessive similarity" — agreement too tight to have come from independent random sampling.

In this vault Simonsohn is the modern, quantitative anchor of the "suspicious perfection" cluster: the leg closest to being a citable *method* rather than a one-off historical verdict. His forensics operationalize with statistics the same inference [[entity-r-a-fisher|Fisher]] applied informally to [[entity-gregor-mendel|Mendel's]] peas — real independent processes carry noise, so its absence is a signature of a defect in provenance.

## References

- [[claim-simonsohn-fabrication-flagged-via-excessive-similarity-to-random-sampling]] — the "excessive similarity" fabrication-detection method.
- [[claim-chiou-2013-coin-size-study-retracted-for-excessive-similarity]] — a concrete case he caught and saw retracted (p<.000025).
- [[observation-suspicious-perfection-independence-absence-signals-defect]] — the general law his method formalizes.
- [[moc-too-clean-to-be-real]] — the data-forensics map that cross-links his method as the "too clean" test's home.