LLM learners are younger, more likely male, and score higher on technology affinity than non-learners¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r1–r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2030: baselines and conditions change

Subclaims¶

q2 i? LLM learners were significantly younger than non-LLM learners (M = 31 vs M = 35) and more likely to be male (53% vs 64% female among avoiders). → Terzimehić 2025 q2 i? LLM learners scored significantly higher on the Affinity for Technology Interaction scale (M = 3.92) than those who avoided LLMs for learning. → Terzimehić 2025 (2)

Evidence¶

Terzimehić 2025¶

Terzimehić, N., Bühler, B., & Kasneci, E. (2025). Conversational AI as a Catalyst for Informal Learning: An Empirical Large-Scale Study on LLM Use in Everyday Learning. https://arxiv.org/abs/2506.11789 doi:10.48550/arxiv.2506.11789

q2 · i? · associational · r1

ANOVA comparing age across three groups (non-LLM users, non-LLM learners, LLM learners) in the survey of 776 participants. The text reports "LLM learners tend to be younger on average (𝑀 = 31,𝑆𝐷 = 9.57), whereas those who did not use LLMs for learning were older (𝑀 = 35,𝑆𝐷 = 11)." An ANOVA confirmed significance (F = 6.68, p < 0.01).

"LLM learners tend to be younger on average (𝑀 = 31,𝑆𝐷 = 9.57), whereas those who did not use LLMs for learning were older (𝑀 = 35,𝑆𝐷 = 11) on average"

Terzimehić 2025 (2)¶

Terzimehić, N., Bühler, B., & Kasneci, E. (2025). Conversational AI as a Catalyst for Informal Learning: An Empirical Large-Scale Study on LLM Use in Everyday Learning. https://arxiv.org/abs/2506.11789

q2 · i? · associational · r2

Survey-based comparison of Affinity for Technology Interaction scores across groups. LLM learners scored highest (M = 3.92, SD = 0.55, α = 0.71), significantly above avoiders (p < 0.01). No standardised effect size is printed.

"LLM-learners scored highest on the ATI scale (𝑀 = 3.92.𝑆𝐷 = 0.55, 𝛼 = 0.71 out of 5), significantly above those who avoided using LLMs (𝑝 < 0.01)"

Discussion¶

Learner Variables¶

• Gender — predictor: outcomes differ by group: a gap reflecting circumstances and treatment, not a cause

Related Claims¶

• Gender, age and English proficiency show nuanced group differences in chatbot attitudes and concerns — related
• Female students are significantly more likely than male students to use AI for schoolwork and to seek advice from AI tools — reports the opposite
• Male students report significantly higher overall technological self-efficacy than female students (p = 0.001) — related
• Female gender, higher autonomous motivation, lower prior knowledge, and higher final course grades each predict more frequent voluntary StatBot use among graduate statistics students — reports the opposite
• 88% of survey respondents have incorporated LLMs into their everyday learning practices
• Affinity for technology interaction differed across clusters, with text reproducers highest and integrators scoring higher than reconstructors (d = 0.59) — related
• Male respondents were significantly more likely than non-male respondents to have no concern about their labour being used for military purposes — related