Daniel Quin

Date of Submission



Due to the short- and long-term benefits associated with better student engagement teachers, parents, and the broader community have sought to understand how best to improve students’ engagement. Teachers and modifications to teaching, as a proximal factor to students’ engagement, have been a focus of efforts to enhance engagement. Furthermore, ecological models of adolescent development assert that factors from the individual student, family, peer group, and community are likely to be influential in the development of adolescent outcomes, including student engagement. To date the empirical engagement literature has failed to consider the potential for factors from multiple contexts to concurrently influence student engagement. Further, studies of student engagement have been limited in that they have prioritised either traditional school-based indicators of engagement (i.e., academic grades, suspension, attendance, dropout) or students’ subjective perspectives of engagement (i.e., psychological). These student perspectives are in response to the demands of school and encompass: i) a student’s overt and less readily observable behaviours; ii) liking and enthusiasm for school; and iii) efforts to understand the prescribed curriculum.

The aims of the research were to: i) understand the extent to which teachers can improve students’ engagement; ii) elaborate upon non-teaching factors that influence engagement and; iii) recommend teaching modifications that can be employed to improve students’ engagement.

Three studies were undertaken to address the overall research aims. The first study was a systematic review of existing research. Thirty-three cross-sectional and 13 longitudinal studies that investigated associations between teacher-student relationships and multiple indicators of engagement were synthesised. The second study, a self-report survey of 88 Year 7 students in Victoria, Australia, conceptualised teaching via self-determination theory. The students reported their perceptions of their teachers, family support, and individual factors including academic grades and mental health. Finally, a pre-existing dataset, from the International Youth Development Study was analysed. Adolescents (n = 719) were surveyed in Grade 10 and again in Grade 11 on a range of factors from the individual, school, family, peer, and community contexts.

In Study 1 (Chapter 6), results indicated that when students and teachers formed a high quality caring relationship students were more likely to have better academic grades, attendance, and psychological engagement. Students were also more likely to have reduced levels of disruptive behaviours, suspensions, and dropout. Study 2 (Chapter 8) presented results of hierarchical regression analyses. After controlling for individual and family factors, better quality teaching was uniquely associated with behavioural and emotional engagement. The discussion explored the need for an integrated model of teaching to improve students’ engagement. The third and final study identified limits to teaching modifications within an ecological model of adolescent engagement. Specifically, Grade 10 teacher support was not a statistically significant predictor of Grade 11 engagement when factors from the individual, family, peer, and community were introduced to the analyses. It was proposed that prior individual educational experiences (i.e., academic grades and engagement) exerted a greater influence on high school engagement than short-term teaching modifications. Overall, teaching remains a proximal factor to students’ engagement. Students who experience high-quality caring relationships with their teachers and better classroom instruction and management were more likely to have better engagement. This was the case across both school-based and students’ subjective indicators of engagement. However, limits to teaching exist. Prior educational experiences and factors from the family, peer, and community were statistically significant predictors of engagement in the ecological models presented.

Future research into engagement would benefit from longitudinal study designs that collect student-reported and school-based indicators of engagement at multiple time-points across the duration of a student’s academic career. Moreover, theoretical models of engagement should seek to elaborate upon the likely bi-directional relationships between the current predictors and student engagement outcomes. Practical recommendations included a need to recognise the limits to teaching and a need to identify how precisely a student’s engagement can be supported.


School of Psychology

Document Type


Access Rights

Open Access


295 pages

Degree Name

Doctor of Philosophy (PhD)


Faculty of Health Sciences