Students predominantly conceptualize the LLM as an Assistant rather than Peer or Expert, and role distributions differ significantly across assignments¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r1 · q3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Assistant was the most common LLM role among users across all assignments, most visible in Algorithms (61.6%) and Web Development (63.0%). → Nizamani 2026 q3 i? The distribution of roles (assistant, peer, expert, none) differed significantly across assignments (χ2 = 358.73, df = 18, p < 0.001). → Nizamani 2026 (2)

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 role conceptualization (assistant, peer, expert, none) reported by 211 students per assignment. The article reports Assistant as most common, with majorities in "Algorithms (61.6%) and Web Development (63.0%)"; Expert use ranged from 0.0% (Software Engineering) to 9.0%.

"Across all assignments, Assistant usage was the most common LLM role among students who used LLMs. This pattern is most visible in Algorithms (61.6%) and Web Development (63.0%), where a majority of students reported using LLMs in an assistive role."

Nizamani 2026 (2)¶

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

Chi-square test of independence on the role distribution across the seven assignments (N=211 per assignment). The article reports "χ 2 = 358.73, df = 18, p < 0.001"; no effect size is printed.

"A chi-square test of independence confirmed that this distribution differed significantly across assignments (χ 2 = 358.73, df = 18, p < 0.001)."

Discussion¶

Related Claims¶

• Role framing shifts toward non-use in team contexts, with non-use more than doubling — related
• Motivational orientation is strongly associated with the role students assign to GenAI: intrinsically motivated students most often position it as an instructor, extrinsically motivated students as a replacement tool — related
• Perceived GenAI role is systematically associated with metacognitive dimensions activated: replacement-tool users show no evaluation or elaboration, collaborator users show predominantly higher-order activity — related
• Prompt-count distributions differ significantly across assignments, with moderate prompting most common and extended prompting (10+) rare except in Algorithms (18.0%) — related
• The detection gain comes from restoring the missing prompt-response relation, not merely assigning an assistant role: the full prefix outperforms the empty prompt by 14.73% and 12.57% AUROC on Qwen2.5-3B and 5.63% and 5.33% on Llama-3.1-8B — related
• LLM adoption varies substantially across CS subfield assignments under a shared instructional intervention, with highest use in Algorithms (89.6%) and Web Development (80.6%) and lowest in Software Engineering (15.2%) — related
• Students issue fewer prompts on team assignments, a medium-large within-student difference — related
• A reflection-structured assignment produced predominantly Strategic Dialogue content, offering preliminary, non-causal evidence that instructional framing may shift path distribution — related