This systematic review addresses a gap that sits at the heart of the belonging literature: while research consistently shows that sense of school belonging is associated with positive academic, social and mental health outcomes, comparatively little rigorous work has examined what schools can actually do to build it. The authors searched seven electronic databases and the Cochrane Central Register from 1999 to February 2021, ultimately identifying 22 controlled trials — 14 randomised controlled trials and 8 quasi-experimental studies — that evaluated school-based interventions targeting belonging in adolescents aged approximately 12 to 18.
Before turning to the interventions themselves, the paper addresses the persistent definitional problem in this field. “School belonging” is used interchangeably in the literature with school connectedness, school bonding, school membership and school engagement, and this inconsistency creates genuine difficulty in comparing findings across studies. The authors sidestep the terminology debate by anchoring their inclusion criteria to measurement: only studies using psychometric instruments aligned to Goodenow and Grady’s (1993) definition — feeling appreciated, respected, supported and involved in the school social environment — were included. They also apply Fredricks et al.’s (2004) three-dimensional framework of behavioural, cognitive and emotional engagement as a lens for categorising what each intervention actually targeted.
The 22 studies spanned 13 countries (predominantly the US, with three from Australia) and involved a total of 13,834 students across 159 schools. Interventions were diverse in design: most were delivered in person and in group settings, ranged from two days to three years in duration, and addressed varying combinations of the three belonging domains. Fourteen of the 22 studies reported positive effects on school belonging.
Several patterns emerged across the effective interventions. Almost all targeted the behavioural domain — social skills training, prosocial behaviour development, bullying reduction and peer relationship building feature consistently as active ingredients. Emotional domain strategies, including mindfulness, emotional regulation, mentoring and building positive teacher-student relationships, were also prominent in effective programs. Cognitive approaches — growth mindset framing, self-reflection, identifying character strengths and adaptive thinking — were present in several successful interventions, particularly where the program sought to shift students’ self-perceptions in relation to school. The paper notes that effective interventions typically combined strategies across multiple domains rather than targeting one in isolation, which aligns with the view that belonging is itself multidimensional and multisystemic.
Vulnerable student groups — those at risk of academic failure, students with disabilities, students requiring high levels of mental health support — showed positive responses to belonging interventions in the five studies that specifically targeted these populations, though the authors note this evidence base is too small to draw firm conclusions.
Despite the weight of positive evidence, the review is candid about the field’s limitations. The heterogeneity of interventions makes it impossible to identify which specific components, formats or durations drive improvement. Statistical analysis found no significant associations between intervention attributes (group vs individual, in-person vs online, duration) and outcomes, largely because the evidence base is too small and varied. The authors also flag that very few of the 22 studies were designed with the primary purpose of building belonging — many treated it as a secondary or complementary outcome — which means the evidence on intentional belonging-building interventions remains thin