An exploratory study of the relationship between student distress and learning behaviours in the post-pandemic context : a learning analytics approach

Abstract

Students’ mental wellbeing is of paramount concern at higher education institutions (HEIs) worldwide. Their mental health issues have deteriorated following the COVID-19 pandemic. Distress is emotional sufferings characterised by depression and anxiety, sometimes accompanied by somatic symptoms such as fatigue, insomnia and ache (Viertiö et al., 2021). There is a pressing need to track student distress and identify those mentally at-risk for intervention at HEIs, but the current methods posit methodological pitfalls compromising their effectiveness. Addressing these challenges, this study examined the relationship between student distress and learning behaviours in learning management systems (LMS), exploring methodological alternatives for distress detection using a student-centred learning analytics approach. Based on the control-value theory of achievement emotions (Pekrun, 2006) and self-regulated learning models (Winne & Hadwin, 1998; Zimmerman, 2002), this study was conducted during the COVID-19 pandemic amongst 300 first-year undergraduate students from different subject areas in their first semester of higher education at a university in Macao, a southern city in China. Their activity traces collected from all the courses they enrolled in LMS during the semester were analysed in association with their self-report distress. The findings provided empirical evidence in support of the theoretical assumption that connects distress with students’ self-regulated learning behaviours in LMS. Early distress detection from LMS traces was also found possible. As the study was conducted in the default teaching and learning settings of the university, the findings have implications regarding the development of affective learning analytics for real-life practices. They also provide insights for enhancing institutional practices safeguarding students’ mental wellbeing.