
|
Transforming Peer Assessment in Higher Education: An AI-Supported Approach Using the Peerscholar Platform
Abstract
This study is devoted to a comprehensive analysis of peer assessment platforms with the support of artificial intelligence (AI) and empirical substantiation of the effectiveness of implementing the PeerScholar system in the process of professional training of future mathematics teachers. In the theoretical part of the work, a systematic comparison of nine leading intellectual platforms (RiPPLE, EduPCR, Peerceptiv, Kritik, PeerScholar, EvaluMate, Gradescope, Peergrade, ALEKS) was carried out, which allowed to identify the most relevant and economically affordable solutions for Ukrainian higher education institutions in conditions of limited resources. The empirical component of the study included 56 third-year students of the specialty "Secondary Education (Mathematics)", of which 53 respondents (n=53) participated in the final assessment, having completed five full cycles of peer evaluation of methodological developments in geometry during the semester. The use of a mixed-methods research design made it possible to combine quantitative data analysis of the questionnaire structure (32 points on the Likert scale, α=0.89) with qualitative thematic analysis of open-ended responses and in-depth interviews. Statistical results demonstrated a high level of overall satisfaction with the platform (M=4.28, SD=0.62), with the methodological aspect of using the system receiving the highest priority in the perception of students (M=4.36, SD=0.61). It was found that future teachers showed a significantly higher interest in constructive feedback (M=4.81, SD=0.42) than in receiving grades (M=4.42, SD=0.52). A significant increase in inter-rater reliability of assessments from rs=0.51 to rs=0.71 over the semester was found, which indicates a progressive development of the assessment competence of future teachers. The study proves that the integration of AI frameworks not only optimizes the workload of the teacher, but also acts as a catalyst for deep methodological reflection, allowing students to effectively identify their own gaps in knowledge through the prism of critical analysis of the work of their classmates.
Keywords
Peer Assessment
Artificial Intelligence
Higher Education
Mathematics Teacher Training
Peerscholar
Methodological Competence.
References
|
Proceedings of the International Conference on Applied Innovations in IT
by
Anhalt University of Applied Sciences
is licensed under
CC BY-SA 4.0
·
This work is licensed under a
Creative Commons Attribution-ShareAlike 4.0 International License
All works are licensed under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0), unless otherwise noted.
Published by ICAIIT in cooperation with Anhalt University of Applied Sciences.