Using relationship-focused reflection to improve teacher-child relationships and teachers' student-specific self-efficacy.

The present study evaluated whether LLInC (Leerkracht-Leerling Interactie Coaching in Dutch, or Teacher-Student Interaction Coaching), a teacher-based coaching-intervention, yielded improvements in dyadic affective teacher-child relationships in elementary school (Grades 2-6). Based on attachment theory, LLInC aims to foster more flexible and differentiated mental representations of teachers' relationships with individual children with whom they experience relationship difficulties. Using a quasi-experimental design, we compared an intervention group of teachers (n = 46 teachers and 92 children) receiving LLInC with a control group receiving no form of intervention (n = 32 teachers and 88 children). To investigate possible transfer effects, we asked teachers from the intervention group to report on their relationships and self-efficacy beliefs regarding two other children with whom they experienced relationship difficulties as well (n = 46 teachers and 81 children). Multilevel models were used to examine intervention effects on teachers' perceptions of relationship quality (i.e., Closeness, Conflict, Dependency), and teachers' student-specific self-efficacy beliefs for Behavior Management and Emotional Support. Teachers receiving LLInC reported short-term improvements in Closeness and self-efficacy beliefs for Emotional Support and decreases in Conflict as compared to control teachers. Similar improvements in Closeness and self-efficacy for Emotional Support were found for the intervention-transfer group as compared to control teachers. Also, teachers receiving LLInC had short-term and longer-term improvements in self-efficacy beliefs for Behavior Management as compared to control teachers. These improvements regarding Behavior Management were not found for the intervention-transfer group.