EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI as an evaluation instrument: Examining handwritten solutions in the context of mathematical representations

Hilmi Karaca

RIED-Revista Iberoamericana de Educación a Distancia · 18.03.2026 · 29 (2)

DOI: 10.5944/ried.47084 ↗

Kurz gefasst

Fachartikel in der Zeitschrift RIED-Revista Iberoamericana de Educación a Distancia, online erschienen am 18.03.2026. Verfasst von Hilmi Karaca. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid integration of generative artificial intelligence (GenAI) into education necessitates a re-evaluation of evaluation practices, particularly for complex tasks such as handwritten mathematical solutions involving multiple representations. This study evaluates the performance of an artificial intelligence (AI) based tool (ExEvAI) in evaluating handwritten responses involving symbolic, table, and graphic representations. Employing a mixed-methods explanatory sequential design, data were collected from 89 undergraduate mathematics education students and evaluated by both the AI and human experts. Quantitative analyses revealed a strong, positive correlation between AI and human scores, indicating high consistency in ranking students. However, the AI demonstrated a significant "positive scoring bias," grading more leniently than experts in symbolic and table questions, whereas no significant difference was found in graphic representations. Qualitative analysis revealed that the AI demonstrated strict adherence to the evaluative prompts, producing highly transparent and structured scoring justifications. Furthermore, the model exhibited strong pedagogical potential by effectively applying partial scoring and providing precise diagnostic feedback on student errors. These findings suggest that AI can serve as a valuable "co-evaluator" to reduce instructor workload and provide detailed feedback. However, due to observed scoring biases, the presence of the human factor in the evaluation process is still important to ensure reliability. The study recommends developing hybrid evaluation models where AI supports, rather than replaces, the pedagogical judgment of educators in complex evaluations.

Zitieren

APA

Karaca, H. (2026). AI as an evaluation instrument: Examining handwritten solutions in the context of mathematical representations. RIED-Revista Iberoamericana de Educación a Distancia, 29(2). https://doi.org/10.5944/ried.47084

Vancouver

Karaca H. AI as an evaluation instrument: Examining handwritten solutions in the context of mathematical representations. RIED-Revista Iberoamericana de Educación a Distancia. 2026;29(2). doi:10.5944/ried.47084

BibTeX
@article{Karaca2026,
  author = {Karaca, Hilmi},
  title = {AI as an evaluation instrument: Examining handwritten solutions in the context of mathematical representations},
  journal = {RIED-Revista Iberoamericana de Educación a Distancia},
  year = {2026},
  volume = {29},
  number = {2},
  doi = {10.5944/ried.47084}
}
Kostenlos starten →