AI-enhanced teaching, learning and assessment for student recruitment and retention in financially constrained English universities

About the Project

English universities face dual pressures: a deepening financial crisis and the rapid integration of artificial intelligence (AI) into teaching, learning and assessment (TLA). With rising operational costs, frozen tuition fees and declining international income, many institutions are projected to operate in deficit. At the same time, student use of AI tools is now near‑universal, yet institutional strategies remain uneven. Although existing studies explore AI’s pedagogical implications, no research has examined how AI‑enhanced TLA can support recruitment and retention under financial constraint, revealing an urgent gap for policymakers and sector leaders.

This PhD project investigates how AI‑enabled TLA can be designed and governed to enhance the attractiveness of English higher education. It examines what students value in AI‑supported provision, how expectations align with national and institutional policies, and which AI interventions can measurably improve enquiry‑to‑application conversion, offer‑holder yield, satisfaction and retention. The project connects educational innovation with economic sustainability, offering evidence to guide universities navigating financial pressures.

The research adopts a mixed‑methods, interdisciplinary design across three instrumental case studies representing a post‑92, a red brick/plate glass and a Russell Group university. Data collection includes a large‑scale student survey (UG to doctoral), post‑2022 national and institutional policies analysed using thematic analysis, and evaluations of AI prototypes such as adaptive tutoring, assessment copilots and AI‑based student support.

Economics‑related methods form a core analytical component. Recruitment outcomes—including conversion and offer‑holder yield—will be assessed through econometric modelling, enabling robust estimation of AI’s measurable effects. A difference‑in‑differences (DiD) design will compare outcomes between AI‑intervention and non‑intervention contexts, strengthening causal inference. To determine financial feasibility, incremental cost‑effectiveness ratios (ICER) will estimate the relationship between intervention costs and improvements in recruitment and retention. Retention and progression will be modelled using institutional datasets, integrating both educational and economic outcomes.

The project will deliver an AI‑enhanced TLA framework and an indicative economic model for sector strategy. Beneficiaries include students, admissions teams, senior leaders and policymakers seeking scalable, cost‑effective approaches to sustaining English higher education.

Supervisors

Yung-Lin Wang

Dr Iryna Kushnir