Students' preferences for AI versus human scoring and feedback vary significantly across assessment scenarios, with strong human preference for high-stakes tasks¶

Claim · All claims
Evidence · 1 study · associational r1 · q2 · i1 small
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i1 A chi-square test of independence showed a significant association between scenario and preferred scoring/feedback method, with a small-to-moderate printed effect size (Cramér's V = 0.22). → Yildirim-Erbasli 2026 q2 i? Human scoring and feedback were strongly preferred for high-stakes tasks such as the final paper worth 40% of the course grade, while AI-only was least preferred in most scenarios. → Yildirim-Erbasli 2026

Evidence¶

Yildirim-Erbasli 2026¶

Yildirim-Erbasli, S. N., Ilgun Dibek, M., Thomas, M. L., & Lesoway, N. (2026). When Do Undergraduate Students Prefer AI? Insights into AI Scoring and Feedback. Behavioral Sciences, 16(7), 1196. https://doi.org/10.3390/bs16071196

q2 · i1 · associational · r1

Chi-square test of independence across ten assessment scenarios in the survey of 93 undergraduates. The association was significant, "χ2(27) = 134.57, p < 0.001"; the printed effect size was Cramér's V = 0.22, a small-to-moderate association, coded i = 1. Descriptively, 76 participants preferred human scoring and feedback for the final paper (40% of course grade).

"The results indicated a significant association between scenario and preferred scoring/feedback method, χ2(27) = 134.57, p < 0.001. The effect size, as measured by Cramér's V, was 0.22, indicating a small-to-moderate association between assignment context and scoring/feedback preferences."

Discussion¶

Related Claims¶

• Undergraduate students' preferences for AI feedback characteristics differ significantly from an equal distribution for tone, length, format, and focus, but not for timing — related
• Preferred scoring and feedback methods differ significantly across assignment types, with AI-based scoring rarely preferred and mixed options most chosen for low-stakes assignments — related
• Frequency of AI use predicts AI feedback preference, while comfort with technology and frequency of human feedback predict relative preference for AI over human grading
• Post-interaction reflections reveal four preference themes: AI as formative learning aid, preference for human authority, conditional acceptance, and skepticism about educational impact — related