Spellcheck and autocorrect are not foolproof: efficacy is bounded near 80% and the most common student writing error is now correctly spelled but inappropriate words¶

Claim · All claims
Evidence · 1 study (2 entries) · review r2–r3 · q2
Currency · newest study 2021 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In a corpus analysis of over 3000 student papers, the overall error rate was stable across decades (2.45 errors per 100 words in 2008 vs 2.26 in 1988), but misspellings fell to 5th place while correctly spelled but incorrect words became the most common error, comprising nearly 14% of all mistakes. → Steven C. Pan 2021 q2 i? The upper bound of spellcheck efficacy in some studies is about 80%, leaving roughly 20% of spelling errors for the writer to address. → Steven C. Pan 2021 (2)

Evidence¶

Steven C. Pan 2021¶

Steven C. Pan, Timothy C. Rickard, Robert A. Bjork. (2021). Does Spelling Still Matter— and If So, How Should It Be Taught? Perspectives from Contemporary and Historical Research. Educational Psychology Review. https://doi.org/10.1007/s10648-021-09611-y

q2 · i? · review · r2

The review reports Lunsford and Lunsford's (2008) corpus analysis of student papers: overall error rate "had not substantially changed (i.e., 2.45 errors per 100 words in 2008, versus 2.26 and 2.11 errors per 100 words in 1988 and 1917"), with correctly spelled but incorrect words (e.g., homonyms) now the most common error at nearly 14% of mistakes.

"a follow-up study by Lunsford and Lunsford ( 2008) found that the overall rate of writing errors (of all types) had not substantially changed (i.e., 2.45 errors per 100 words in 2008"

Steven C. Pan 2021 (2)¶

Steven C. Pan, Timothy C. Rickard, Robert A. Bjork. (2021). Does Spelling Still Matter— and If So, How Should It Be Taught? Perspectives from Contemporary and Historical Research. Educational Psychology Review. https://doi.org/10.1007/s10648-021-09611-y

q2 · i? · review · r3

The review reports that "in some studies, the upper bound of spellcheck efficacy has been about 80%", citing MacArthur et al. (1996) and others, leaving the remaining errors to the writer. The review attributes this to the cited studies.

"in some studies, the upper bound of spellcheck efficacy has been about 80% (e.g., MacArthur et al., 1996"

Discussion¶

Related Claims¶

• AI scoring consistency approaches or exceeds 90% for most rubric items, but spelling is replaced by a deterministic LanguageTool-based approach achieving 68% accuracy with 100% consistency — related
• LLM-based scoring of Spanish argumentative writing achieves 60–80% per-item accuracy, slightly below human inter-rater agreement — related
• Only 155 of 2,005 misspellings occur in words on the Simplified Spelling Board's first recommended lists — related