Social interactions within the classroom are fundamental in shaping students’ academic performance, social inclusion, and overall school climate. However, traditional educational assessments focus mainly on individual traits, often neglecting the relational dynamics that shape learning and student well-being. In Chile, the Agencia de Calidad de la Educación (ACE)—the government institution responsible for assessing education quality through national and international evaluations—recognizes the importance of integrating methodologies that capture these relational aspects to enhance the continuous improvement of schools and classrooms.

This study explores the impact of friendship networks and social dynamics in the classroom using Social Network Analysis (SNA). Integrating relational measures into school climate assessments aims to generate empirical evidence to inform pedagogical strategies and public policies aligned with the objectives of the ACE.

To achieve this, a pilot study will be conducted in 100 Chilean schools, focusing on students in 8th and 12th grades. The study will analyze trust, classroom cohesion, and peer interactions using survey-based data collection. Statistical modeling and network analysis will examine how social network structures relate to academic performance, student well-being, and school inclusion.

The expected findings will provide valuable insights into how different network structures influence key school outcomes, such as academic achievement and classroom well-being. This initiative represents an innovative approach to understanding social dynamics in education and fostering more inclusive school environments. Since the study is still in its early stages, we will present our progress at the conference to foster discussion and guide its implementation.

Institutions seek to shape the behaviors of their constituents. Because individuals are embedded in complex webs of social relations, by shaping personal behaviors, institutions may unwittingly reshape network structure. In this paper, I introduce the concept of relational externalities — the unintended relational consequences of a policy or intervention. I elaborate two different forms that relational externalities commonly take, the mechanisms that bring them about, and the role they play in creating or amplifying inequalities. I then empirically illustrate this concept through the case of the punishment of smoking behavior in a sample of U.S. high schools. I show that harsh punishment of smoking behavior is associated with the social isolation of smokers and increased homophily on smoking status in friendship networks. A subsequent exploration of the mechanisms behind these effects reveals that they are likely driven by non-smokers eschewing friendships with smokers. Finally, I show that relational externalities of anti-smoking policies are most acute for students with the least behavioral elasticity with regards to smoking. Students who are least able to change their behaviors change their friends, putting them at a dual disadvantage.

Research Practice Partnerships (RPPs) are now prominent collaborative structures in the Research and Development infrastructure of education systems. These long-term working arrangements focus on key dilemmas and challenges facing practitioners. Importantly, they are mutualistic and include carefully designed rules, roles, routines, and protocols that structure interaction to support school improvement.

This presentation will cover the efforts of a three-year RPP between researchers at the University of Glasgow and practitioners working at a local secondary school as they attempted to implement an approach to professional learning known as practitioner enquiry with all school staff. Practitioner enquiry is a cyclical process of professional learning where educators identify problems of practice and engage with relevant research and embedded evidence from their own classrooms to inform changes in their teaching.

This presentation will explore how Social Network Analysis can be used to evaluate the inter-organisational engagement strategy (e.g. the roles and routines for structured interaction) between researchers and school-based educators working to maintain an RPP for both knowledge exchange and to support the adoption of a new approach to professional learning. Furthermore, the network analysis was used to inform changes to the engagement strategy to strengthen the inter-organisational links between the University and secondary school as the implementation matured. Specifically, the network analysis encouraged the research team to engage with formal middle leaders at the secondary school instead of the senior leadership team. Implications for knowledge mobilisation and fidelity in the programme implementation will be discussed.

In this study, we expand our prior longitudinal SNA research in Russian vocational schools. Along with the drinking and smoking behavior of adolescents, we include a wide range of risk behaviors and explore the dynamics of peer effects on sexual behavior and sexual literacy. Unlike smoking which is highly visible, and alcohol consumption, which is relatively visible among friends, sexual behavior is highly private. Still, we expect that the time of sexual debut may exhibit both effects of influence and selection within the years of vocational education and training. Moreover, our qualitative research shows that sexual knowledge is predominantly shared informally in peer communication networks, and thus we expect to find influence effects in the dynamics of sexual (il)literacy. We address these questions using a longitudinal dataset collected between 2016 and 2020 from 13 vocational colleges in St. Petersburg, Russia. Students in vocational schools often come from low SES families, and risk behaviors are more prevalent there than in academic schools, making this environment advantageous for our study. The survey includes data on smoking, drinking, and sexual experience across four waves, along with measures of sexual literacy obtained in two waves. We employ RSiena for statistical modeling to investigate peer influence and selection dynamics. In the presentation, we will discuss our preliminary results as well as the limitations of our data and analysis in the hope of constructive feedback.

Rural science teachers, often grappling with extreme professional isolation due to significant geographic separation and a lack of same-subject teaching colleagues nearby, face unique challenges. This longitudinal study, conducted in a Western US state, delves into the impact of technology-mediated lesson study (TMLS) professional development on rural chemistry and biology teachers' collaboration, advice-seeking, and friendship networks. Using a stepwise approach to specify stochastic actor-oriented models (SAOMs) in RSiena, we analyzed changes in the three networks over three years. The survey respondents included teachers who participated in TMLS across two cohorts (N=13) and those who did not (N=67). Our findings reveal that, while overall collaboration networks expanded initially and then stabilized, the advice-seeking and friendship networks exhibited continuous growth, indicating the positive impact of TMLS. Outdegree (density), reciprocity, and triadic closure were significant in all three networks, indicating the cohesive formation of subgroups. Despite minimal homophily effects, geographic separation emerged as a significant barrier to tie formation. However, TMLS program participation significantly increased tie formation in all three networks, even across long distances, effectively overcoming geographic barriers to professional connection. These findings underscore the value of sustained technology-enhanced professional development models that foster long-term collaboration, trust, and knowledge exchange among rural teachers.

Research on social norms shows that targeting perceptions of inclusive and diverse norms can be a fruitful method for promoting positive intergroup relations. However, despite this promising earlier research, it is important to better understand how social norms that promote positive intergroup relations can be spread effectively throughout the wider community. Individuals within social networks whose behaviors strongly affect perceptions of social norms, referred to here as social referents, may be important to engage in interventions to effectively spread information and innovations within social networks. Building on this notion, we evaluated our school intervention from the INCLUSIVITY project (www.inclusivitynorms.com). We implemented two waitlist-controlled intervention trials in five high schools (n = 3,911, aged 10-19); two of these schools were at high risk for conflicts between ethnic and religious groups. RSiena was used to examine intervention effects in school networks and to control for confounding processes (e.g., friendship influence). In intervention schools, social referents increased their friends’ respect and tolerance toward discriminated groups and also enhanced perceived inclusive and diverse social norms, even after controlling for friendship influence. No evidence of outsized influence from social referents was found in control schools. Furthermore, evidence for school-level changes was limited and inconsistent. Thus, targeting social referents may be a promising strategy to promote tolerance and positive intergroup relations among friends, yet it has limited potential to change intergroup tolerance at the school level.

Powering Up is a novel school-based intervention that utilizes three sources of power to combat peer victimization in middle school: (1) Why Power is a 12-lesson curriculum designed to alter the maladaptive attributions of at-risk victims; (2) Friend Power uses experimental techniques to build friendships between victims and influential peers in their grade; and (3) Peer Power harnesses the power of these influential youth to change peer norms about the acceptability of bullying. The current research utilizes longitudinal social network analysis (SNA) approaches such as Siena and Latent Space Modeling to examine diffusion and peer influence effects among Powering Up participants in three U.S. middle schools (N=1004). One notable finding from Peer Power was that closeness to peer leaders in the network significantly increased participants’ endorsement of defending behaviors. Various other applications of these dynamic tools will be explored using pre- and post-intervention data from Powering Up, and the role of SNA in conducting peer-led interventions will be discussed.