Behavior Science-Informed Dropout Early Warning Systems in El Salvador, Guatemala, and Honduras

Group Submission Type: Formal Panel Session

Proposal

In the Northern Triangle countries of El Salvador, Guatemala, and Honduras, addressing educational dropout remains a crucial challenge for achieving universal education and equitable development. This panel explores innovative Dropout Early Warning Systems (DEWS) designed to identify and mitigate dropout risks at both primary and secondary education levels. DEWS utilize a range of metrics and data analytics to detect early warning signs—such as attendance, behavior, and academic performance, collectively known as the ABCs—allowing educators to proactively support at-risk students and address contributing factors to dropout (Bruce, 2011).

Effective DEWS not only enhance student retention and completion rates but also support holistic development by integrating tailored school and community-based interventions. These systems provide timely, actionable data that strengthen school culture and policies, ensuring that students receive the necessary care and resources to overcome challenges. By addressing psychosocial needs and following students through critical transitions, DEWS have the potential to foster an inclusive educational environment where all students have the opportunity to succeed. Additionally, improved educational outcomes can lead to better employment prospects, potentially reducing the primary drivers of irregular migration.

The success of DEWS relies on teachers and schools consistently entering and responding to data on student performance and behavior. However, previous research (UNESCO, 2022) has highlighted barriers to effective DEWS implementation, such as complex user interfaces, delays in data processing, and inadequate support for data entry and response protocols.
Key discussion questions for this panel include:
• What behavior science-informed design features—such as routines, defaults, reminders, and incentives—can improve DEWS user experiences and take-up?
• How have new digital tools enhanced teachers’ and system leaders’ ability to identify learning gaps and dropout risks?
• What level of anonymity is sufficient to address teacher accountability concerns while informing system-wide needs?
• Can improved digital inclusion, performance monitoring, and online learning access contribute to system resilience in the face of disruptions?

The panel will present three distinct DEWS models and practices, each informed by behavioral science principles. Key components examined include data collection methodologies, data visualization, predictive analytics, and intervention strategies. The presentations will highlight how localized contextual factors—such as socioeconomic conditions and educational infrastructure—shape the effectiveness of these systems. The discussion will culminate with reflections on the potential and limitations of scaling and adapting DEWS frameworks to promote sustainable educational improvements and reduce dropout rates across the region.

References:

Bruce, M., Bridgeland, J. M., Fox, J. H., and Balfanz, R. 2011. On Track for Success: The Use of Early Warning Indicator and Intervention Systems to Build a Grad Nation. Washington, DC: Civic Enterprises.

UNESCO 2022. Early warning systems for school dropout prevention in Latin America and the Caribbean. UNESCO Regional Bureau for Education in Latin American and the Caribbean. https://unesdoc.unesco.org/ark:/48223/pf0000380354_eng

Sub Unit

• Latin America SIG

Chair

• Amber K Gove, RTI International

Individual Presentations

• Incorporating Behavior Science to Improve Early Warning System Usage and Increase Retention in El Salvador - Beatriz Slooten Navarro, CUBIC/Save the Children International

• But Will It Make My Job Easier? Early Warning Systems Development with and for Teachers in Guatemala - Amber K Gove, RTI International; Cynthia Carolina del Aguila, BEQTRTI