Statlint reads the statistics in any research paper and mathematically proves which reported results cannot be real — seven peer-reviewed forensic methods, unified into a single report.
Papers report means no dataset could produce, standard deviations that break algebra, and p-values that contradict their own test statistics. Almost none of it gets checked — reviewers and editors have neither the time nor the tooling.
Paste a results paragraph or upload a PDF, JSON, or CSV. Processed in memory, never stored.
Seven forensic checks recompute and cross-examine every number the paper reports.
Proven impossibilities first, heuristic flags separated, each explained in plain English.
Five return mathematical proofs of impossibility. Two return heuristic flags. Statlint never conflates the two.
Statlint draws a hard line between a mathematical impossibility — proven, the numbers cannot all be right — and a statistical suspicion, a flag worth a second look that proves nothing on its own. The engine is validated against thousands of real datasets to guarantee it never flags genuine data. Every finding is a question to put to the authors, not a verdict on them.
The methods are field-agnostic by construction — if a paper reports means, SDs, p-values, or test statistics, Statlint can check it.
Paste a results paragraph or upload a file. Your report opens in a new tab.
Analyzed in memory. Documents are never written to disk or retained.