Manual reasoning is the most consistently used verification method across assignments, with no significant difference in its use across subfields¶

Claim · All claims
Evidence · 1 study · associational r1 · q3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2031: baselines and conditions change

Subclaims¶

q3 i? Manual reasoning ranged from 70.7% (Networking) to 82.4% (HCI) across assignments, and a chi-square test showed no significant difference across assignments (χ2 = 5.73, df = 6, p = 0.455). → Nizamani 2026

Evidence¶

Nizamani 2026¶

Nizamani, S. B., Lee, Y., Donekal Chandrashekar, N., Ellis, M., & Ramakrishnan, N. (2026). Understanding Student Use of Large Language Models Across Computer Science Subfields. https://arxiv.org/abs/2610.01158 doi:10.48550/arxiv.2610.01158

q3 · i? · associational · r1

Descriptive analysis of verification strategies among students who reported active LLM use (assistant, peer, or expert role) across the seven assignments. The article reports manual reasoning "ranging from 70.7% to 82.4% across assignments"; a chi-square test showed no significant difference (χ2 = 5.73, df = 6, p = 0.455); equivalence was not tested.

"manual reasoning was the most consistently used verification method, ranging from 70.7% to 82.4% across assignments. HCI showed the highest manual reasoning rate (82.4%), while Networking had the lowest (70.7%)."

Discussion¶

Related Claims¶

• Verification strategy type varies by assignment context: testing is more common in structured tasks (Algorithms 87.8%) while web search is more common in open-ended tasks (HCI 56.0%) — related
• Verification is nearly universal among LLM users, and most students combine multiple verification strategies rather than relying on a single one — related