Scientific Publication Credential Authority Failure Through Peer Review of Methodology Without Data Verification at Diederik Stapel Case

Diederik Stapel was a prominent social psychologist who held professorships at the University of Amsterdam, the University of Groningen, and Tilburg University over a career spanning two decades. His research produced striking findings on topics including stereotyping, discrimination, and the influence of environmental factors on social behavior. His papers appeared in leading journals including Science, Journal of Personality and Social Psychology, and Psychological Science. His work was widely cited and earned him prestigious awards and leadership positions within the field.

Stapel fabricated data. He invented datasets, manufactured experimental results, and in many cases never conducted the experiments described in his published papers. He typically controlled the data personally, telling co-authors and PhD students — who contributed to study design, analysis, and writing — that he had collected the data himself or through research assistants, and providing them only with processed datasets rather than raw data. His co-authors wrote papers, earned PhDs, and built careers on findings derived from data that did not correspond to any actual experiment or observation.

In August 2011, three young researchers in Stapel's department noticed statistical patterns in his data that appeared too consistent — effect sizes that were unusually uniform, results that were too clean, and datasets that lacked the variability characteristic of real experimental data. They reported their concerns to a senior colleague, who escalated to the university administration. Tilburg University suspended Stapel and convened an investigative committee. Within weeks, Stapel confessed to fabricating data.

Three Dutch universities jointly investigated the scope of the fraud. The Levelt Committee, named for its chair, psycholinguist Willem Levelt, published its report in November 2012. The investigation identified at least 55 publications containing fabricated or manipulated data. The committee found that Stapel had been fabricating data since at least 1996 — spanning his tenures at three universities — and that the fabrication had intensified over time as he became more senior and faced less institutional scrutiny of his work. Fifty-eight papers were ultimately retracted from the journals that had published them.

Peer review — the system by which scientific publications are credentialed — evaluated Stapel's papers for methodological soundness, statistical appropriateness, and the logical connection between data and conclusions. This evaluation confirmed that the papers were internally consistent: the methodology described was appropriate for the research questions posed, the statistical tests applied were correct for the data types presented, and the conclusions followed from the results. What peer review did not and structurally could not evaluate was whether the data existed — whether the experiments were conducted, whether participants were recruited, whether the measurements were taken. The publication credential certified the paper's internal logic. It did not certify the paper's connection to reality.

The Levelt Committee's investigation found broader structural conditions that enabled the fraud. Stapel operated in a field where data sharing was not standard practice — researchers were not routinely required to make raw data available to reviewers, co-authors, or the field. He controlled the data personally and restricted access to processed results. Junior collaborators who depended on him for career advancement were not in a position to demand access to raw datasets. The institutional review and promotion processes evaluated his publication record — the number and prestige of his publications — without independently verifying the work those publications described. His rising institutional status reduced rather than increased scrutiny, creating the condition seen across the collection: institutional value functioning as insulation from verification.

Stapel was stripped of his PhD by Tilburg University in 2012 — an unprecedented action in Dutch academia. He was barred from the university and reached a settlement with prosecutors that included community service rather than criminal prosecution. Fifty-eight papers were retracted. The case prompted significant reform discussions in social psychology and science broadly, including the movement toward mandatory data sharing, pre-registration of studies, and open science practices designed to make raw data and analysis scripts available for independent verification. Several journals and funding agencies implemented data availability requirements as conditions of publication or funding.