Bias transfer is asymmetric: prompt condition affects agency in female-target essays but not male-target essays, suppressing female agency while male writing remains stable¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q2 · i2 medium
Currency · newest study 2026 · checked against its sources 2026-10-16 (automated, 2 of 2 entries) · not reviewed by a person

Subclaims¶

q2 i2 A significant Gender×Condition interaction (F=4.41, p=.014, η²p=.070) confirms the biased prompt widened the gender gap selectively rather than depressing agency uniformly. → Ariyan Hossain 2026 q2 i? Prompt condition significantly affected female-target agency (F(2,60)=4.71, p=.013) but not male-target agency (F(2,57)=2.16, p=.125); no effect size is printed for the simple effects. → Ariyan Hossain 2026 (2)

Evidence¶

Ariyan Hossain 2026¶

Ariyan Hossain, Kazi Kamruzzaman Rabbi, Farig Sadeque, S M Taiabul Haque. (2026). Contaminated Collaboration: Measuring Gender Bias Transfer in LLM-Assisted Student Writing. arXiv preprint. https://arxiv.org/abs/2606.15914 doi:10.48550/arxiv.2606.15914

q2 · i2 · causal · r2

Two-way ANOVA (N=123) predicting agency score by biography subject gender and prompt condition. The interaction effect was F=4.41, p=.014, η²p=.070; biography gender was the strongest predictor (F=20.17, p<.001, η²p=.147).

"The two-way ANOV A (Table 4) revealed a signif- icant Gender× Condition interaction alongside main effects of gender and prompt condition."

Ariyan Hossain 2026 (2)¶

Ariyan Hossain, Kazi Kamruzzaman Rabbi, Farig Sadeque, S M Taiabul Haque. (2026). Contaminated Collaboration: Measuring Gender Bias Transfer in LLM-Assisted Student Writing. arXiv preprint. https://arxiv.org/abs/2606.15914

q2 · i? · causal · r2

Simple-effects analysis within the 123-participant experiment. Female agency was significantly lower in the biased condition than neutral (p=.023) and control (p=.037); control and neutral did not differ (p=.982). The article prints F and p values for these simple effects but no effect size.

"Prompt condition significantly affected agency in female-target essays ( F(2,60) = 4.71 , p= .013) but not male-target essays (F(2,57) = 2.16 , p=.125 ), with all male pairwise contrasts non- significant (allp>.10 )."

Discussion¶

Learner Variables¶

• Gender — moderator: an instructional effect differs by group
• Stereotype Threat — moderator: an instructional effect differs with it

Related Claims¶

• AI systems are non-neutral socio-technical artifacts embedding cultural assumptions and power relations — the general claim this finding bears on, with the studies behind it
• Students assisted by a gender-biased LLM produce career-plan essays with a significantly larger male-female agentic gap than students in neutral-LLM and no-AI control conditions — related
• A gender-biased system prompt induces large gender-differentiated language in LLM-generated career essays, while a neutral prompt does not — related
• Students exposed to the biased LLM showed no awareness of its influence, reporting lower perceived AI influence than neutral-condition participants despite larger bias effects — related
• Biased LLM assistance increases students' gender-stereotypic occupational recommendations: 71.4% of biased-condition essays recommended a gender-stereotypic occupation versus 45.0% (control) and 39.0% (neutral) — related
• Female students reported higher learning agency than male students after controlling for GenAI literacy and year of study — related
• A neutrally configured LLM assistant produced the lowest rate of gender-stereotypic occupation suggestions, below even the no-AI control condition — related
• Gender prompts shift feedback stance: female students receive affective, first-person, empathy-oriented feedback while male students receive objective, task-oriented critique — related
• Learners perceived male PALs more positively and recalled more with them than with female PALs, reflecting gender stereotypes — related