Student’s perceptions of teaching quality are important for understanding teaching effectiveness, setting research agendas, and fostering positive teacher-student interactions (Kikas & Magi, 2017; Wallace et al., 2016). Despite recognizing the importance of students' perceptions, a significant gap exists in the literature concerning how adolescents, specifically those at-risk perceive teacher emotional support, particularly in both academic and vocational tracks. Given the heightened sensitivity of students at-risk, there is a critical need to investigate how students at-risk status and teacher emotional support influences their perceptions in these educational contexts (Murray & Greenberg, 2001; O’Connor, 2010; Roorda et al., 2011, 2017). This study aims to fill this gap by investigating adolescent students’ perceptions of teacher emotional support and examining the impact of teacher emotional support and students at-risk in both academic and vocational groups.

The study employs the theoretical underpinnings of both the bioecological model of human development and the process-person-context-time model (PPCT), which emphasize the interconnectedness of various systems (Bronfenbrenner & Morris, 1979, 2006). Delving into the educational context, the Teaching through Interactions framework (Hamre et al., 2013; Hofkens & Pianta, 2022) adapts these principles as it underscores the significance of the classroom as a context where proximal processes, such as teacher-student interactions unfold. Furthermore, the microsystem of the classroom is conceptualized, highlighting that the characteristics of both teacher (i.e., teacher emotional support), and students’ (i.e., at-risk status) play a pivotal role in shaping the quality of teacher-student interactions and students’ perceptions of these interactions (Bronfenbrenner & Morris, 2006; Pianta et al., 2003).  

The study utilized data from the initial time-point (T1) of a mixed-methods cluster randomized controlled trial (RCT) in upper secondary schools, known as "INTERACT" which examines the impact of a video-based coaching intervention on teacher-student interactions (Ertesvåg et al., 2022). The sample included 1341 students and 98 teachers in Norway from both vocational and academic tracks. Teachers, participating in a web-based survey before randomization of the intervention reported on emotional support provided to students without specific student details. Students, recruited through their respective teachers at the start of the school year, participated in a web-based survey assessing their perceptions of emotional support from their designated "INTERACT" teacher during a regular class lesson. Student recruitment and consent were conducted ethically, approved by the Norwegian Agency for Shared Services in Education and Research with reference number 210803. Student-perceived emotional support was measured using a revised scale capturing trust, respect, and interest in the teacher-student relationship (Bru et al., 2022; Tvedt et al., in progress). At-risk status was identified through a comprehensive approach involving students reporting an Individualized Education Plan (IEP) during lower secondary school and an achievement score below a 2.6 grade point average (Hoen et al., 2019). Gender was obtained from registered data (0=Male, 1=Female), while SES was measured using parents' highest education levels (1=Compulsory school; 2=Upper secondary education; 3=College or university). Both were used as control variables. Teacher-reported emotional support was measured through a scale assessing individual perceptions of emotional support (Ertesvåg et al., 2011). Teachers' work experience (1-5, 6-10, 11-14, 15+ years) and educational qualification (1-5) were used as control variables. Given the hierarchical nature of the data, where individual students were nested within classrooms, and the research focus was to investigate differences or similarities between vocational and academic groups, a doubly latent multigroup multilevel structural equation modelling was applied to evaluate the measurement and structural model hypothesizing a positive association between teacher emotional support and student-perceived emotional support, controlling for teacher-related variables at the classroom level, and a negative association between at-risk and student-perceived emotional support, controlling for student-related variables across both academic and vocational groups (Marsh et al., 2009, 2012). Descriptive analyses used IBM SPSS Statistics (Version 29), while Mplus 8.10 (Muthén and Muthén, 1998-2023) was employed for other analyses. Model fit was assessed using various criteria, with cutoff values indicating good fit

Preliminary analyses using intraclass coefficient (ICC) to assess the impact of grouping students into vocational or academic tracks on student perceived teacher emotional support showed revealed significant differences between the two groups, underscoring the role of group membership and justification for multilevel modelling (Hox, 2013). Furthermore, preliminary analyses of the measurement model invariance testing indicated that students’ perceptions of teacher emotional support are consistent both within and between classrooms, and across academic and vocational tracks. The optimal fitting model was the configural model, which was freely estimated, ensuring valid comparisons between the two groups (Marsh et al., 2012). Additionally, all standardized factor loadings were statistically significant at p < .001). Finally, preliminary analyses for the structural model revealed that in both vocational and academic groups, students at-risk perceived lower levels of emotional support from their teaching. In the vocatonal group, teacher emotional support did not align with how students perceived their teachers as being emotionally supportive. However, in the academic group, teacher emotional support did align with student perceived emotional support. In conclusion, the study contributes valuable insights into the complex dynamics of teacher-student interactions, with a particular focus on students at-risk in different educational tracks. The findings have implications for educational practices and policy.

Hamre, B. K., R. C. Pianta, J. T. Downer, J. DeCoster, A. J. Mashburn, S. M. Jones, J. L. Brown, E. Cappella, M. Atkins, and S. E. Rivers. 2013. “Teaching Through Interactions: Testing a Developmental Framework of Teacher Effectiveness in Over 4,000 Classrooms.” The Elementary School Journal 113 (4): 461–487. https://doi.org/10.1086/669616. Hofkens, T. L., and R. C. Pianta. 2022. “Teacher–Student Relationships, Engagement in School, and Student Outcomes.” In Handbook of Research on Student Engagement, 431–449. Springer International Publishing. https://doi.org/10.1007/978-3-031-07853-820 . Ertesvåg, S. K., G. S. Vaaland, and M. K. Lerkkanen. 2022. “Enhancing Upper Secondary students’ Engagement and Learning Through the INTERACT Online, Video-Based Teacher Coaching Intervention: Protocol for a Mixed-Methods Cluster Randomized Controlled Trial and Process Evaluation.” International Journal of Educational Research 114: 102013. https://doi.org/10.1016/j. ijer.2022.102013 . Bronfenbrenner, U., & Morris, P. A. (2006). The bioecological model of human development. In W. Damon & R. M. Lerner (Series Eds.) & R. M. Lerner (Vol. Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (6th ed., pp. 793–828). Hoboken, NJ: Wiley & Sons. Roorda, D. L., H. M. Y. Koomen, J. L. Spilt, and F. J. Oort. 2011. “The Influence of Affective Teacher– Student Relationships on Students’ School Engagement and Achievement: A Meta-Analytic Approach.” Review of Educational Research 81 (4): 493–529. https://doi.org/10.3102/ 0034654311421793 . Pianta, R. C., B. K. Hamre, and J. P. Allen. 2012. “Teacher-Student Relationships and Engagement: Conceptualizing, Measuring, and Improving the Capacity of Classroom Interactions.” In Handbook of Research on Student Engagement, 365–386. Springer. https://doi.org/10.1007/978-1-4614-2018-717 .