Student LLM usage patterns varied across data science courses: Course-BD showed the highest frequency and over-reliance with near-zero perceived learning effect¶

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

Subclaims¶

q2 i? Course-BD students reported the highest usage frequency and over-reliance with a perceived learning effect near zero, while Course-ML students reported lower usage and a positive learning effect. → Yuan An 2026

Evidence¶

Yuan An 2026¶

Yuan An, Lei Wang. (2026). Student Use of LLMs and the Limits of AI-Generated Question Difficulty in Data Science Courses. https://arxiv.org/abs/2609.27063 doi:10.48550/arxiv.2609.27063

q2 · i? · associational · r1

Descriptive cross-course comparison on the final survey wave (Course-BD n=7, Course-ML n=3, Course-AML n=4) found Course-BD mean use frequency 3.43 and over-reliance 4.00 versus Course-ML's 2.67 and 2.33; learning effect means were 0.14 versus 1.00.

"Course-BD showed the highest usage frequency, the highest self-reported over-reliance, and a perceived learning effect near zero (Figure 5)."

Discussion¶

Related Claims¶

• LLM usage varied across assignments and tasks, increasing for later assignments (especially discussion points), while Low-Reliance students concentrated use early and reduced it after Midterm 2
• 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%)
• Students' perceived deep-learning benefits of LLMs declined over the term while satisfaction and time savings remained high
• High-frequency LLM users show higher minimum course outcomes and reduced score variability, though interaction level is not a dominant outcome predictor — related