Why Analysts Spend 80% of Their Time Cleaning Bad Data

Who created and checked this page

Correction history

Every accepted correction to this page is recorded with the exact change, so readers can see how the page improved over time.

1. 29 July 2026 · corrected by Emir Baycan

   Corrected a fabricated citation in the article's account of the October 2020 UK Public Health England Excel row-limit incident: the widely-reported '125,000' impact figure was falsely attributed to a University of Nottingham professor's letter to the British Medical Journal (no such letter exists). The real source of the 125,000 figure is Fetzer and Graeber's academic study estimating additional COVID-19 infections (not 'contacts not notified') resulting from the delayed contact tracing.

   Before

   Article attributed a specific real-sounding statistic to a specific named expert and publication venue that could not be verified as the actual source.

   After

   Corrected the attribution to the real source (Fetzer and Graeber's study) and the real metric it measured (additional infections, not contacts not notified), while preserving the accurate core statistic of 15,841 dropped cases.

   Why: A real number from a real study was reattributed to a different real expert and a fabricated publication venue, the same fabrication pattern found across this batch of legacy articles.

   View the full record →

Full contribution timeline

1. Correction 29 July 2026

   Emir Baycan: Corrected a fabricated citation in the article's account of the October 2020 UK Public Health England Excel row-limit incident: the widely-reported '125,000' impact figure was falsely attributed to a University of Nottingham professor's letter to the British Medical Journal (no such letter exists). The real source of the 125,000 figure is Fetzer and Graeber's academic study estimating additional COVID-19 infections (not 'contacts not notified') resulting from the delayed contact tracing.

   Proof record →