Learning Effectiveness Checklist

Contributions

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

1. 30 July 2026 · corrected by Emir Baycan

   Corrected unverified statistics across five citations and a minor omission in the sixth, including a wrong journal for a real Pooja Agarwal study and a wrong journal/title/metrics for a Duolingo efficacy study.

   What the page claimed

   Article omitted 'interleaved practice' from Dunlosky et al.'s real moderate-utility tier. It included an unverified specific '26 percent higher' prediction statistic attached to a real Kornell & Bjork 2008 study (the real finding is that 71% judged massed practice superior versus only 12% correctly judging spaced practice superior, not a percentage-point prediction gap). It misattributed Pooja Agarwal's real classroom research to the wrong journal (stated Psychological Science; the real paper is in Educational Psychology Review) and included an unverified statistics ('0.6 grade levels,' 'eight months') not found in the real study. It unverified statistics for the National Training Center at Fort Irwin ('40 to 60 percent better,' an unverified 1996 TRADOC report with an '18 months in 30 days' figure) that could not be verified in any real source. It converted Mathematica Policy Research's real KIPP findings (framed in the real study as 'months of additional learning') into a different, unsupported metric ('percentile points') and included an unverified '2,900 students' sample size. And it misattributed Duolingo's real Cindy Blanco with the wrong title (real: Senior Learning Scientist, not 'chief learning officer'), the wrong journal (real: Foreign Language Annals, not 'Bilingualism: Language and Cognition'), an unverified sample size (4,100), and an entirely unverified 2019 internal study with a specific '49 percent' vocabulary-retention figure.

   What was corrected

   Dunlosky's moderate-utility tier corrected to include interleaved practice. The Kornell/Bjork statistic replaced with an accurate qualitative description of the real, more dramatic finding (71% vs 12%). Agarwal's citation corrected to the real journal, with the included an unverified statistics replaced by accurate general descriptions of the real study's scale and durability. The NTC passage rewritten to describe the real underlying principle without the unverified statistics and nonexistent report. The KIPP passage corrected to use the real 'months of learning' metric rather than the unverified percentile-point conversion, with the unverified sample size removed. The Duolingo passage corrected to Blanco's real title and the real journal, with the unverified sample size and 2019 internal study replaced by an accurate general description of Duolingo's real published usage-data findings.

   Why: This continues the established pattern for this batch: nearly every citation anchored to a real researcher, institution, or company, but almost every specific statistic was unverified, altered, or misattributed to the wrong publication - only Dunlosky's high and low utility tiers were fully accurate as originally stated, with the moderate tier needing just one addition.

   View the full record →

Contributors In This Page