Designing a Conceptual Framework for AI-Based Adaptive Curriculum Planning with Emphasis on Learning Analytics and Students’ Individual Differences: An Exploratory Mixed-Methods Study
Learning, Training, and Education in Schools in the Third Millennium · 2025 · 2 (4) · 1-26
DOI: 10.61838/jltestm.104 ↗Kurz gefasst
Mixed-Methods-Studie in der Zeitschrift Learning, Training, and Education in Schools in the Third Millennium, 2025. Verfasst von Ali Radmehr.
Abstract (Originalsprache)
This study aimed to design and validate a conceptual framework for AI-based adaptive curriculum planning with emphasis on learning analytics and students’ individual differences. An exploratory sequential mixed-methods design was employed. In the qualitative phase, 20 experts in curriculum planning, educational technology, artificial intelligence in education, educational measurement and assessment, and learning analytics in Tehran were selected through purposive sampling. Data were collected through semi-structured interviews and analyzed using thematic analysis. Based on the qualitative findings, a researcher-developed questionnaire was constructed and its face and content validity were examined. In the quantitative phase, the questionnaire was administered to 400 upper-secondary students in Tehran selected through multistage cluster sampling. Quantitative analyses included internal consistency reliability, composite reliability, average variance extracted, and confirmatory factor analysis. Model fit was assessed using χ²/df, GFI, AGFI, CFI, TLI, NFI, IFI, RMSEA, and SRMR. The qualitative analysis identified 10 major components: data-driven curriculum infrastructure, learning analytics, identification of individual differences, learner-profile modeling, adaptation of learning objectives and instructional content, adaptation of teaching strategies and learning activities, intelligent assessment and continuous feedback, AI-based decision-making, teacher supervision, and ethical, security, and supportive requirements. In the quantitative phase, standardized factor loadings ranged from 0.67 to 0.91 and Cronbach’s alpha coefficients ranged from 0.86 to 0.92. Composite reliability values ranged from 0.89 to 0.94, while average variance extracted ranged from 0.61 to 0.71. Model-fit indices indicated satisfactory fit: χ²/df=2.19, GFI=0.91, CFI=0.95, TLI=0.94, NFI=0.92, IFI=0.95, RMSEA=0.055, and SRMR=0.046. The findings demonstrated that AI-based adaptive curriculum planning is a multidimensional, dynamic, and data-driven system in which learning analytics and individual differences provide the basis for dynamic learner profiling and intelligent adaptation of curriculum components. Teacher supervision and ethical safeguards function as essential regulatory mechanisms. The acceptable psychometric properties and model-fit indices support the proposed framework as a valid foundation for the development of intelligent and personalized curriculum-planning systems.
Zitieren
APA
Radmehr, A. (2025). Designing a Conceptual Framework for AI-Based Adaptive Curriculum Planning with Emphasis on Learning Analytics and Students’ Individual Differences: An Exploratory Mixed-Methods Study. Learning, Training, and Education in Schools in the Third Millennium, 2(4), 1-26. https://doi.org/10.61838/jltestm.104
Vancouver
Radmehr A. Designing a Conceptual Framework for AI-Based Adaptive Curriculum Planning with Emphasis on Learning Analytics and Students’ Individual Differences: An Exploratory Mixed-Methods Study. Learning, Training, and Education in Schools in the Third Millennium. 2025;2(4):1-26. doi:10.61838/jltestm.104
BibTeX
@article{Radmehr2025,
author = {Radmehr, Ali},
title = {Designing a Conceptual Framework for AI-Based Adaptive Curriculum Planning with Emphasis on Learning Analytics and Students’ Individual Differences: An Exploratory Mixed-Methods Study},
journal = {Learning, Training, and Education in Schools in the Third Millennium},
year = {2025},
volume = {2},
number = {4},
pages = {1--26},
doi = {10.61838/jltestm.104}
}
EUVIMED