Burnout and job satisfaction represent two important aspects of teachers’ occupational well-being, which is of particular interest with regard to increasing teacher shortage worldwide (OECD, 2005; UNESCO Institute for Statistics, 2016), high attrition rates (den Brok et al., 2017; Lindqvist et al., 2014), and frequent incapacity to work due to mental and physical illness (Seibt et al., 2009; Simone et al., 2016). From a theoretical perspective, teachers’ occupational well-being is thought to affect the quality of teacher-student interactions and students’ cognitive-motivational development (Jennings & Greenberg, 2009). For instance, according to the broaden-and-build theory (Fredrickson, 2001), high levels of occupational well-being (e.g., job satisfaction) likely foster individual resources (e.g., time, energy, empathy, care), which enable teacher to create positive, effective and cognitively activating learning environments, promoting student motivation and successful learning. In turn, teachers experience high levels of emotional exhaustion likely have less emotional, social and cognitive resources available for accomplishing the various tasks of the teaching profession (Bakker & Costa, 2014; Hobfoll, 1989; Maslach & Leiter, 1999). Likewise, teachers’ well-being might directly affect students through emotional contagion (Frenzel et al., 2021; Oberle & Schonert-Reichl, 2016; Tikkanen et al., 2021). Numerous cross-sectional studies documented this association (Arens & Morin, 2016; Gastaldi et al., 2014; Klusmann et al., 2022). However, longitudinal evidence is scarce and focused on individual paths of the hypothesized mediating process (Shen et al., 2015).

Therefore, this study aimed to investigate data of N = 45 teachers and their students (N = 806) from a three-wave longitudinal study to test the mediating role of teacher- and student-reported quality of teacher-student interactions (i.e., emotional support, classroom management and instructional support) for the predictive effect of teachers’ emotional exhaustion and job satisfaction – two key aspects of teachers’ occupational well-being – on students’ motivation and achievement. In doing so, we extend previous research not only by testing the complete meditation model longitudinally, but also by considering multiple dimensions of teachers’ occupational well-being and accounting for a wide variety of interaction- and student outcomes.

Preliminary analyses revealed significant association consistent with our expectations between teachers’ preceding well-being, in terms of job satisfaction and emotional exhaustion, and student-rated quality of classroom management and instructional support at the following measurement point. For teachers’ self-report only, cross-sectional relationships emerged between teachers’ well-being and classroom management. While no significant relationship emerged between teachers’ well-being and students’ outcomes, the preliminary analyses suggested a significant positive link between emotional support and classroom management at the previous measurement point and subsequent student motivation (i.e., utility value and self-concept) for both teachers’ self-report and student ratings. For students’ achievement, significant positive associations were only found for teachers rated emotional support, classroom management and instructional support.

According to the pre-registration (https://osf.io/npk8c), we will conduct full longitudinal mediation models to investigate whether teacher- and student-rated quality of teacher-student interactions (emotional support, classroom management, and instructional support) at the second measurement point mediates the relationship between teachers’ occupational well-being (i.e., emotional exhaustion and job satisfaction) in the beginning of the school year and students’ motivation and achievement at the third measurement point (Little et al., 2009). Because our data has a multi-level structure (i.e., students nested within teachers), we will address this structure in our analyses (Preacher et al., 2010; Raudenbush & Byrk, 2002). In doing so, we will control for prior levels of the quality of teacher-student interactions (at least for teachers’ self-reports) and students’ outcomes.

The findings will be discussed with regard to theoretical and practical implications, study limitations, and avenues for future research.

Educationalists, psychologists, and economists have agreed that social-emotional characteristics are essential to students’ successful life outcomes (Duckworth & Yeager, 2015; Heckman & Kaut, 2012). The core domains of social–emotional characteristics can be organized based on three psychological needs in the Self-Determination Theory (SDT): (a) task-orientation (competence), (b) self-orientation (autonomy), and (c) other-orientation (relatedness) (Ryan & Deci, 2000; Schoon, 2021). Based on the contextual aspect of the SDT, social-emotional characteristics might not contribute to different outcomes uniformly as different outcomes are associated with distinct goals and life experiences (Ryan & Deci, 2000). This leads to the question of how social-emotional characteristics are related to two important outcomes — academic and well-being outcomes — for secondary school students. Also, whether the predictions of social-emotional characteristics are similar between general and school-related academic and well-being outcomes.

This conceptualization of social-emotional characteristics can map into main components in the Expectancy-Value Theory (EVT) to explain the predictions of social-emotional characteristics on school-related and general academic outcomes (Eccles et al., 1983). The EVT posits that expectancy beliefs, values, general self-schema and socialization are linked to school-related and general academic outcomes. Expectancy beliefs (self-orientation domain) and value (task-orientation domain) directly influence school-related and general academic outcomes; general self-schema and socialization processes have more distant influences on school-related and general academic outcomes.

Regarding the predictions of social-emotional characteristics on well-being, during the period of secondary school, students are particularly in need of strong competence, autonomy and relatedness to foster their development (Eccles et al., 1993; Erikson, 1968). Moreover, given that establishing a coherent self-identity is an important developmental task for students, social interactions with others play an important role in helping students shape their identities, which leads to better well-being (Eccles et al., 1993; Erikson, 1968). Thus, social-emotional characteristics in the self-orientation and other-orientation domains might strongly predict general and school-related well-being.

So far, no longitudinal study investigated the joint predictions of social-emotional characteristics from multiple domains on school-related and general academic and well-being outcomes (Guo et al., 2023; Soto et al., 2022). Thus, we investigated two main research questions: (1) the longitudinal predictions of social-emotional characteristics from the task-orientation, self-orientation, and other-orientation domains on school-related and general academic outcomes, and (2) the longitudinal predictions of social-emotional characteristics from the task-orientation, self-orientation, and other-orientation domains on school-related and general well-being outcomes.

To answer our research questions, we analyzed 4,744 students who were followed from grade 9 to grade 10. Predictors were social-emotional characteristics in three domains in grade 9: (1) task-orientation domain: learning motivation, (2) self-orientation domain: academic self-concept and self-esteem, and (3) other-orientation domain: empathy and social anxiety interaction. Academic outcomes were academic performance (school-related outcome) and expectations of studying at university (general outcome) in grade 10. Well-being outcomes were school and life satisfaction in grade 10, corresponding to school-related and general well-being. Control variables were immigrant status, family socioeconomic status, gender, school track, prior cognitive ability and well-being. Multivariate linear regression and probability analyses were applied.

The results showed that learning motivation, academic self-concept, and social anxiety interaction positively predicted academic performance. Learning motivation and academic self-concept positively predicted expectations of studying at university. Moreover, self-esteem and empathy significantly positively predicted life satisfaction, but social anxiety interaction negatively predicted life satisfaction. However, only learning motivation positively predicted school satisfaction. In sum, the results supported the EVT that task-orientation and self-orientation domains were strongly related to school-related and general academic outcomes. These findings additionally supported the contextual aspect of the SDT that the predictions of social-emotional characteristics are rather domain-specific.