An Evidence-based Model for Integrating Artificial Intelligence into the Academic Quality Management Systems in South African Private Higher Education Institutions

Abstract

The purpose of this article is to present the quantitative validation phase of a sequential exploratory mixed-methods study investigating the integration of Artificial Intelligence into Academic Quality Management Systems in South African Private Higher Education Institutions (PHEIs). This phase empirically tests the conceptual model established in the prior phase through literature review and qualitative thematic analysis. The objective is to determine whether the proposed model functions as a robust measurement structure when assessed against actual data, and to refine it where necessary. The research utilised a survey instrument developed around eight principal themes, including AI's role in strategic planning, teaching and learning, ethics, environmental impact, and interdisciplinary collaboration. Survey responses were analysed to assess whether the model’s constructs demonstrate statistical coherence, using established reliability and measurement tests. The results indicate that all constructs were appropriate for factor analysis, and the retained factors exhibited satisfactory internal reliability. Three constructs were retained without modification, while five required item-level refinement. The analysis did not confirm the original conceptual model unchanged; instead, it produced a revised AI QMS integration model with an empirically validated measurement structure. This refined model can be applied in subsequent research to assess institutional AI QMS integration and highlight areas for improvement. The study provides an important first-stage empirical refinement of a theory-driven model developed through a preceding qualitative phase. The objective was not to produce a finalised psychometric instrument but to identify which theoretically derived dimensions demonstrate sufficient coherence for future investigation.