EUVIMEDEUVIMED entdecken
Crossref

Adoption Barriers to AI-Powered Learning Tools Among University Faculty

Hridoy Hosen Khan, Danish Mahmud

US Journal of New Insights in Tech & Education · 15.07.2023 · 1

DOI: 10.69013/pr0829fw ↗

Kurz gefasst

Fachartikel in der Zeitschrift US Journal of New Insights in Tech & Education, online erschienen am 15.07.2023. Verfasst von Hridoy Hosen Khan und 1 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This mixed-methods study investigates the barriers that shape university faculty adoption of AI-powered learning tools, integrating a quantitative survey of 412 faculty members across five disciplines with semi-structured interviews from 18 purposively sampled participants. Machine learning techniques, including a random forest classifier and k-means clustering, were applied to the survey data to identify the strongest predictors of adoption and to segment faculty into distinct adoption profiles. The random forest model classified faculty adoption status with 72% accuracy and an area under the curve (AUC) of 0.75, identifying perceived usefulness, perceived ease of use, and digital literacy as the strongest predictors of adoption, while technology anxiety, limited institutional support, and workload pressure emerged as the most salient barriers. Cluster analysis revealed three faculty segments, Ready Adopters (34%), Cautious Adopters (37%), and Resistant Faculty (29%), each requiring differentiated support strategies. Qualitative interviews further revealed five recurrent barrier themes: time and workload constraints, insufficient institutional training, concerns about academic integrity and data privacy, discomfort with algorithmic opacity, and a perceived mismatch between AI tools and disciplinary pedagogy. Supplementary robustness checks, including a random forest learning-curve analysis and a comparison against logistic regression, confirmed the stability of the reported predictors across modeling choices. The study concludes that faculty adoption of AI-power ed-learning tools is shaped less by the technology itself than by the organizational and psychological ecosystem surrounding its introduction, and it proposes a tiered, discipline-sensitive framework for institutional AI integration.

Zitieren

APA

Khan, H. H., & Mahmud, D. (2023). Adoption Barriers to AI-Powered Learning Tools Among University Faculty. US Journal of New Insights in Tech & Education, 1. https://doi.org/10.69013/pr0829fw

Vancouver

Khan HH, Mahmud D. Adoption Barriers to AI-Powered Learning Tools Among University Faculty. US Journal of New Insights in Tech & Education. 2023:1. doi:10.69013/pr0829fw

BibTeX
@article{Khan2023,
  author = {Khan, Hridoy Hosen and Mahmud, Danish},
  title = {Adoption Barriers to AI-Powered Learning Tools Among University Faculty},
  journal = {US Journal of New Insights in Tech & Education},
  year = {2023},
  pages = {1},
  doi = {10.69013/pr0829fw}
}
Kostenlos starten →