Originality showed a borderline non-significant trend toward higher accuracy on professional than EFL student writing, while Turnitin showed no significant difference¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r1 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? Originality classified all 48 professional texts correctly versus 44 of 48 EFL texts, with a one-sided Fisher's Exact Test p = 0.058 that did not meet conventional significance; Turnitin showed no significant difference (p = 0.50). → Mohammad Hadra 2026 q2 i? Turnitin demonstrated comparable accuracy for EFL student writing and professional writing, with no statistically significant difference. → Mohammad Hadra 2026 (2)

Evidence¶

Mohammad Hadra 2026¶

Mohammad Hadra, Karleen Cambridge and Mostefa Mesbah. (2026). Evaluating the accuracy and reliability of AI content detectors in academic contexts. International Journal for Educational Integrity. https://doi.org/10.1007/s40979-026-00213-1

q2 · i? · associational · r1

Fisher's Exact Test (one-sided) comparing correct classification of 48 professional texts (100% accuracy for Originality) against 48 EFL texts (91.6%). The article reports "a borderline statistically significant trend (one-sided p = 0.058)" that does not reach conventional significance.

"Fisher’s Exact Test indicated a borderline statistically significant trend (one-sided p = 0.058) toward higher accuracy for professional writing. Although the result does not meet the conventional threshold for statistical significance, the directional trend aligns with emerging literature documenting potential detector bias against non-native or less proficient writers."

Mohammad Hadra 2026 (2)¶

Mohammad Hadra, Karleen Cambridge and Mostefa Mesbah. (2026). Evaluating the accuracy and reliability of AI content detectors in academic contexts. International Journal for Educational Integrity. https://doi.org/10.1007/s40979-026-00213-1

q2 · i? · design · r2

Fisher's Exact Test comparing Turnitin accuracy on 48 EFL texts (91.6%) versus 48 professional texts (93.8%); the article reports "no statistically significant difference (Fisher’s Exact Test, p = 0.50)". Accuracy was used as the primary metric because all texts in this subset were human-authored.

"Turnitin demonstrated comparable accuracy for both EFL student writing and professional writing, with no statistically significant difference (Fisher’s Exact Test, p = 0.50)."

Discussion¶

Related Claims¶

• Both detectors achieved substantially lower accuracy on scientific writing than on humanities texts — related
• Originality outperformed Turnitin in overall accuracy and macro-average recall on a balanced EFL-relevant dataset, but both showed limited consistency across classes — related
• Detector accuracy declined significantly as text length increased, with both detectors performing best on short texts — related
• A detector can achieve high aggregate accuracy while producing materially different over-detection rates across author groups — related
• Both detectors performed poorly on hybrid human–AI texts, with Originality showing near-zero recall for the Hybrid class — related