HARNESSING AI FOR ACADEMIC RESILIENCE: A MIXED METHOD STUDY ON POSTGRADUATE STRESS AND WELL-BEING IN SRI LANKAN UNIVERSITIES

Abstract

This study investigates the escalating mental health challenges faced by postgraduate students in Sri Lankan universities, focusing on stress and well-being. The study involved 110 full-time postgraduate students aged 22–35 from the Open University of Sri Lanka and the University of Colombo. Using a mixed-methods approach, quantitative data were gathered through the DASS-21 questionnaire, analyzed with machine learning algorithms and sentiment analysis to identify stress patterns, while qualitative insights were obtained from threeweek interactions with the Wysa chatbot. Results revealed that 68% of participants exhibited moderate to severe stress, with academic pressure (37.5%) and financial concerns (25%) as key contributors. The chatbot intervention reduced anxiety in 65% of participants, with significant improvement in DASS-21 scores (p < 0.001). This study demonstrates AI’s potential as a scalable, culturally adaptable tool for improving mental health support in universities. It highlights the need for long-term studies, integration into institutional frameworks, and consideration of cultural and financial barriers for sustainable adoption.

Ähnliche Arbeiten