AI-Driven Instructional Leadership: Multilevel Modeling of Principals’ Data Literacy, Predictive Decision-Making, and Teaching Quality
Year: 1405
Type: article
Study design: Observational Study
Abstract: A sequential exploratory mixed-methods design was used. The qualitative phase involved interviews with 19 educational experts and principals, analyzed through open, axial, and selective coding. The quantitative phase surveyed 384 teachers and principals using validated questionnaires assessing data literacy, predictive decision-making, and teaching quality. Data analysis included structural equation modeling and multilevel modeling to assess direct and indirect effects. Results showed that principals’ data literacy significantly predicted predictive decision-making (β=0.63) and teaching quality (β=0.27). Predictive decision-making also significantly influenced teaching quality (β=0.48). Mediation analysis confirmed that part of the effect of data literacy on teaching quality operates indirectly through predictive decision-making (β=0.30). Multilevel analysis indicated significant between-school variance in teaching quality. Findings indicate that AI-driven instructional leadership is most effective when principals possess high levels of data literacy and use predictive decision-making to enhance instructional quality. Predictive decision-making represents a key mechanism linking data literacy to teaching outcomes.
DOI: 10.61838/medda.367
Authors: Amir Amiri Roshkhar, Tavakol Dadras
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W7233580345
