Browsing by Author "Ahmed Ikram Meer"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Chatgpt and passive cheating. A quasi-experimental study on ai assisted exam preparation
    (UMT.Lahore, 2025) Muhammad Umer Daraz Cheema; Ahmed Ikram Meer
    This study investigates the intersection of faculty stress, artificial intelligence (AI) adoption in assessment design, and student academic performance through the lens of the Effort Reward Imbalance (ERI) model. Drawing on a quasi-experimental, quantitative research design, data were collected from 130 higher education faculty members and four undergraduate classes across diverse disciplines. The first hypothesis tested whether higher ERI among faculty members predicted increased reliance on AI tools such as ChatGPT for designing assessments. Statistical analysis, including t-tests and regression, revealed a significant positive relationship, indicating that institutional stressors compel faculty to use AI not out of innovation, but as a coping mechanism. The second hypothesis examined whether students who used AI for exam preparation outperformed peers using traditional methods. Across all four subject classes, AI-prepared students demonstrated significaThis study investigates the intersection of faculty stress, artificial intelligence (AI) adoption in assessment design, and student academic performance through the lens of the Effort Reward Imbalance (ERI) model. Drawing on a quasi-experimental, quantitative research design, data were collected from 130 higher education faculty members and four undergraduate classes across diverse disciplines. The first hypothesis tested whether higher ERI among faculty members predicted increased reliance on AI tools such as ChatGPT for designing assessments. Statistical analysis, including t-tests and regression, revealed a significant positive relationship, indicating that institutional stressors compel faculty to use AI not out of innovation, but as a coping mechanism. The second hypothesis examined whether students who used AI for exam preparation outperformed peers using traditional methods. Across all four subject classes, AI-prepared students demonstrated significantly higher scores and moderate to large effect sizes, raising concerns about “passive cheating”—a phenomenon where students gain unintended advantages due to congruence between AI-assisted learning and AI-influenced assessment design. Subgroup analysis further revealed that female faculty, mid-career academics, and those on research tracks reported higher ERI levels. These findings underscore the cascading effects of institutional stress on both teaching practices and student performance, highlighting the urgent need for policy reforms to address workload equity, assessment integrity, and the ethical integration of AI in higher education. Keywords: Effort-Reward Imbalance, AI in Education, Passive Cheating, ChatGPT-Assisted Learning, Assessment Integritntly higher scores and moderate to large effect sizes, raising concerns about “passive cheating”—a phenomenon where students gain unintended advantages due to congruence between AI-assisted learning and AI-influenced assessment design. Subgroup analysis further revealed that female faculty, mid-career academics, and those on research tracks reported higher ERI levels. These findings underscore the cascading effects of institutional stress on both teaching practices and student performance, highlighting the urgent need for policy reforms to address workload equity, assessment integrity, and the ethical integration of AI in higher education. Keywords: Effort-Reward Imbalance, AI in Education, Passive Cheating, ChatGPT-Assisted Learning, Assessment Integrit