The Central America Security Trends Dashboard was developed in Summer 2026 as part of an ongoing project from Jack D. Gordon Institute for Public Policy at Florida International University. Based on comprehensive regional research, this interactive tool is designed to track and visualize the complex, divergent security landscapes across Central America. Recognizing that security is about more than just the reduction of homicides, the Dashboard highlights how different nations confront organized crime and whether states are successfully restoring legitimate territorial authority.

To capture and visualize these developments, the Dashboard incorporates and regularly updates data across three distinct but interconnected security dimensions:

• National Security: This dimension evaluates threats to state authority, territorial control, border integrity, strategic infrastructure, and national stability. It goes beyond conventional warfighting to capture penetration by transnational organized crime, cross-border illicit flows, attacks on state actors, prison system destabilization, and strategic corruption or capture.
• Public Security: Focused on the law-enforcement and public-order domain, this category tracks geographically concentrated crime patterns. It includes data on homicides, robbery, theft, extortion, assault, kidnapping, gang violence, riots, prison disturbances, and illicit seizures.
• Citizen Security: Representing the human-centered dimension of security, this category is distinct from conventional state-centered approaches. It focuses on the protection of people and democratic coexistence by measuring fear, victimization, mobility constraints, perceptions of insecurity, willingness to report crime, and community-level trust and confidence in the state.

The Dashboard focuses its analysis specifically on four countries—Costa Rica, Guatemala, Honduras, and El Salvador—which together reveal the full diversity of Central America's security trajectories. Rather than moving along a single regional path, all four countries face distinct pressures:

• Costa Rica represents a historically stable democracy dealing with a rapid, logistics-linked surge in organized crime and homicide.
• Guatemala illustrates the persistence of extortion, prison-based criminal command, and weak justice institutions testing a democratic reform agenda.
• Honduras highlights a post-exception transition where peak homicides have dropped, yet communities still live under severe criminal governance, territorial fear, and institutional mistrust.
• El Salvador presents the region's most dramatic drop in visible gang violence, though its success relies on a consolidated emergency-security model that raises significant concerns regarding due process, civic space, and institutional risk.

By tracking these unique national realities, the Dashboard illustrates the central challenge facing the region today: finding security models firm enough to confront armed, extortionary criminal structures, yet disciplined enough to preserve institutional legitimacy, due process, and citizen trust.

Accurately measuring the multidimensional nature of security across Costa Rica, Guatemala, Honduras, and El Salvador presents inherent methodological challenges. While official crime statistics provide a baseline, they frequently under count actual incidents of violence and extortion due to under reporting, fear of criminal retaliation, weak institutional trust, and limited independent access to detention and prison data. To capture a deeper, human-centered perspective, the underlying study incorporates focus group evidence; however, this qualitative data reveals mechanisms, perceptions, and lived experiences rather than statistically representative metrics. Furthermore, data collection methodologies vary by country. In El Salvador, for instance, focus groups were not conducted; instead, the assessment relies on expert interviews and open-ended Delphi surveys. While valuable, these expert and stakeholder perceptions are not directly equivalent to neighborhood-level testimonies and must be triangulated with independent sources, official data, and other qualitative evidence. Any future attempts to aggregate these Delphi indicators into a cross-country quantitative index will also require careful normalization, as scale directions vary across the survey questions.

The Dashboard relies on third-party event data and forecasting models that carry their own operational constraints. Incident tracking from the Armed Conflict Location & Event Data (ACLED) project depends heavily on publicly available reports, local source networks, and media coverage. As a result, fluctuations or gaps in the data may sometimes reflect changes in source access, the reporting environment, or the introduction of new monitoring networks rather than actual on-the-ground shifts in violence. ACLED also employs explicitly conservative fatality estimates, which may further contribute to the under counting of casualties.

To project future security trajectories, the Dashboard incorporates ACLED's Conflict Alert System (CAST). While CAST utilizes machine learning and historical reporting data to probabilistically forecast organized violence events at subnational levels, its predictive capability remains inherently bounded by the availability and accuracy of past data and carries statistical uncertainty. Crucially, the political and security landscapes in these four countries in 2026 are fast-moving. Sudden developments—such as elections, Attorney General or judicial transitions, shifts in state-of-exception measures, prison policies, or the release of new homicide data—can rapidly alter national trajectories. Therefore, the integrated Dashboard serves as a dynamic but point-in-time representation, requiring periodic updates to accurately reflect the complex, shifting realities of Central American security.