School-level factors associated with teacher connectedness: a multilevel analysis of the structural and relational school determinants of young people's health.

Background: Conducting research on the antecedents of teacher connectedness (TC) is key to inform intervention and policy that can leverage the public health potential of teachers for young people's well-being. As part of the EU-funded Teacher Connectedness Project, this study aims to examine the contribution of a variety of school-level factors (including type of school, school size, student-teacher ratio, students per class and teacher gender). Methods: Sample consisted of 5335 adolescents aged 11, 13 and 15 years that had participated in the HBSC study in England. Multilevel multinomial regression was used to examine the contributions of sociodemographic and school-level factors to TC. Results: TC was lower in older adolescents and those from less affluent families, but similar in boys and girls. Regarding school-level factors, it was not the size of the school but the ratio of students per teacher which was significantly associated to TC, with higher student-teacher ratio being significantly associated with lower odds of medium-to-high TC. Some differences between mixed and all-girls schools were also found. Conclusions: Health promotion strategies targeting student-teacher relationships need to consider how TC changes by age and SES and give attention to school-level factors, in particular the student-teacher ratio.

School context and health in adolescence: the role of sense of coherence.

The aim of this study was to examine the role of school variables and sense of coherence (SOC) in the explanation of health. For this purpose, data were collected from a representative sample of adolescents, aged 13 to 18 years, selected for the Health Behavior in School-aged Children survey in Spain. Using Structural Equation Modelling, three nested models were tested that represented different hypothesized models of the relationships among school factors, SOC and health. According to goodness-of-fit indices, in the best model, school-related stress and SOC mediated the influence of support from classmates and teachers on health, and SOC had a direct effect on stress. The results supported the idea that a supportive school climate and SOC are relevant to adolescents' health. Furthermore, SOC was the most influential variable; apart from its direct positive effect on health, SOC seemed to condition the degree to which students perceived school demands to be stressful. Specifically, students with a high SOC were less likely to suffer from high levels of school-related stress.