Equity and Bias in AI Educational Tools: A Critical Examination of Algorithmic Decision-Making in Classrooms
Shumaila Farheena, Azhar Abbas Cheemab, Rai Samee Ullahc, Marium Minhas Bandealid
- aMPhil, Department of Education
- bMS Management Sciences, University of Sargodha
- cDepartment of Computer Science and Software Engineering, Grand Asian University Sialkot
- dPhD Public Policy, Deputy Director (T/C), Directorate of Sindh water sector & barrages improvement, GoS
Papers open inside SciRah from the original publisher or repository; SciRah never hosts them.
Abstract
The integration of artificial intelligence (AI) into education raises significant concerns about algorithmic bias and its impact on equity. This study examines the linkage between teachers' perceptions of AI bias and their concerns for fairness in the classroom while also assessing the influence of teaching level and AI familiarity. Through a cross-sectional survey of 270 educators in Punjab, Pakistan, we utilized quantitative methods, including correlation and regression analysis, to test our hypotheses. Our findings reveal a strong positive association between perceived algorithmic bias and threats to educational equity. We found that school teachers were significantly more aware of AI bias than university faculty. Furthermore, greater teacher familiarity with AI tools correlated with a more nuanced understanding of their potential biases. These results underscore the urgent need to address the "black box" nature of educational AI. The study provides empirical evidence that, without careful regulation and comprehensive teacher training, AI-driven tools risk perpetuating and even exacerbating existing educational inequalities, aligning with global concerns about the ethical deployment of AI in sensitive sectors.
TL;DRThe study provides empirical evidence that, without careful regulation and comprehensive teacher training, AI-driven tools risk perpetuating and even exacerbating existing educational inequalities, aligning with global concerns about the ethical deployment of AI in sensitive sectors. (Semantic Scholar)