Final Performance Evaluation of the Regional Citizen Security Project (InfoSegura) January 31, 2019 This publication was produced at the request of the United States Agency for International Development. It was 1 prepared independently by James C. Jones, Team Leader; Francesca Jessup, Governance Expert; and Carlos Carcach, Data Analytics Expert. i ABSTRACT The purpose of this performance evaluation of the USAID-funded Regional Citizen Security Project (InfoSegura) is to document project achievements and challenges to date, inform adjustments for the remainder of project implementation, and help with appropriate planning for future citizen-security work. The evaluation seeks to answer the following three broad evaluation questions (EQs) regarding InfoSegura: 1) the project’s contribution to change the level of quality and timeliness of crime and violence data definition, collection, and recording; 2) the Central American governments’ use of crime and violence data generated through the project as input to effective citizen security￾related policymaking; and 3) the contribution of the creation of partnerships and transparency between government and civil society to increased collaboration in the design of citizen-security policy. Key illustrative findings indicate that InfoSegura, in all three countries, contributed to capacity development within governmental technical units to collect, record, and use crime and violence data for analysis; strengthened the capacities of source institutions to improve quality of data and information-sharing; and contributed to consolidate capacities developed in technical and source institutions to enhance sustainability. The evaluation further confirmed the use, in important instances, of data generated through the project to inform policy decisions on citizen security￾related issues in each of the three countries. Among key recommendations are the need for: the design and implementation of training programs in the fields of official crime statistics, survey sampling, crime measurement scales, crime analysis for crime investigation and evaluation of crime prevention policies; and the strengthening of analytical capacity for evidence-based policymaking within relevant institutions. Front cover photo: Salvadoran woman attending the launch of the National Survey of Violence against Women 2017, presented by El Salvador’s General Directorate of Statistics and Censuses (DIGESTYC) to the Legislative Assembly on April 13, 2018. Credit: Katherine Rogel/Infosegura ii FINAL PERFORMANCE EVALUATION OF THE REGIONAL CITIZEN SECURITY PROJECT (INFOSEGURA) Submitted: January 31, 2019 Prepared for the United States Agency for International Development under AID-OAA-I-15-00024/TOAID-519-TO-16-00002 USAID/El Salvador Contact: Orlando Hidalgo ohidalgo@usaid.gov Prepared by: James C. Jones, Social Scientist and Team Leader Francesca Jessup, Governance Expert Carlos Carcach, Data Analytics Specialist Submitted by: USAID Monitoring, Evaluation, and Learning Initiative Calle Circunvalación #261, Colonia San Benito, San Salvador, El Salvador Tel.: (503) 2423-7486 DISCLAIMER The authors’ views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development or the United States Government. iii CONTENTS Acronyms............................................................................................................................ v Executive Summary ....................................................................................................... viii Evaluation Purpose and Evaluation Questions....................................................................................viii Project Background ..................................................................................................................................viii Design, Methods, and Limitations..........................................................................................................viii 1.0 Evaluation Purpose And Questions ........................................................................... 1 1.1 Evaluation Purpose................................................................................................................ 1 1.2 Evaluation Questions............................................................................................................ 1 2.0 Project Background..................................................................................................... 2 2.1 Problem and Project Context ............................................................................................ 2 2.2 Change Hypothesis and Causal Logic .............................................................................. 3 3.0 Evaluation Methods and Limitations ......................................................................... 3 3.1 Evaluation Methods ............................................................................................................... 3 3.2 Evaluation Limitations........................................................................................................... 4 4.0 Findings, Conclusions, and Recommendations ........................................................ 5 4.1 EQ1 Findings........................................................................................................................... 5 4.1.1 EQ1: To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region?...................................................................................................................................... 5 4.2 EQ1 Conclusions ................................................................................................................ 14 4.3 EQ1 Recommendations .................................................................................................... 16 4.4 EQ2 Findings ........................................................................................................................ 18 4.4.1 EQ2: To what extent have Central American governments used (as planned) crime and violence data—produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions?.................................................................................... 18 4.5 EQ2 Conclusions ................................................................................................................ 23 4.6 EQ2 Recommendations .................................................................................................... 24 4.7 EQ3 Findings ........................................................................................................................ 25 4.7.1 EQ3: To what extent has the creation of partnerships and transparency (CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy?............................... 25 iv 4.8 EQ3 Conclusions ................................................................................................................ 28 4.9 EQ3 Recommendations .................................................................................................... 28 4.10 EQ3 General Recommendations.................................................................................... 28 ANNEXES ........................................................................................................................30 ANNEX I: EVALUATION SCOPE OF WORK................................................................................. 31 ANNEX II: COUNTRY CONTEXT FOR EL SALVADOR, GUATEMALA, AND HONDURAS ....................................................................................................................................................................... 46 ANNEX III: DATA COLLECTION INSTRUMENTS........................................................................ 50 ANNEX IV: INFOSEGURA THEORY OF CHANGE HYPOTHESIS AND RESULTS FRAMEWORK........................................................................................................................................... 52 ANNEX V: EVALUATION MATRIX ................................................................................................... 54 ANNEX VI: SOURCES OF INFORMATION..................................................................................... 56 ANNEX VII: STATEMENT OF DIFFERENCES – INFOSEGURA PROJECT .............................. 64 LIST OF TABLES Table A1: Summary of Findings, Conclusions, and Recommendations................................................ x Table 1: Evaluation Data Sources.................................................................................................................. 4 Table 2: Counts of Victims of Homicide and Extortion by Data Source, January 2015 – September 2018 ...................................................................................................................................................... 7 v ACRONYMS ADS Automated Directives System APJ Spanish acronym for Alianza por la Paz y la Justicia (Alliance for Peace and Justice) ARENA Spanish acronym for Alianza Republicana Nacionalista (Nationalist Republican Alliance) ASJ Association for a More Just Society CEASCI Spanish acronym for Centro Nacional de Estudio y Análisis de Convivencia y Seguridad Ciudadana (Center for the Study and Analysis of Coexistence and Citizen Security) CEMUJER Spanish acronym for Instituto de Estudios de la Mujer (Institute of Women’s Studies) CENISS Spanish acronym for Centro Nacional de Información del Sector Social (National Center for Information on the Social Sector) CICIG Spanish acronym for Comisión Internacional contra la Impunidad en Guatemala (International Commission against Impunity in Guatemala) COE Center of Excellence for Statistical Information on Government, Crime, Victimization, and Justice CONOSE Spanish acronym for Red de Conocimiento sobre Seguridad Ciudadana (Knowledge Network on Citizen Security) CNSCC Spanish acronym for Consejo Nacional de Seguridad Ciudadana y Convivencia (National Council for Citizen Security and Coexistence) DIA Spanish acronym for Dirección de Investigación y Análisis (Research and Analysis Directorate) DIGESTYC Spanish acronym for Dirección General de Estadística y Censos (General Directorate of Statistics and Censuses) ENP Spanish acronym for Estrategia Nacional de Prevención del Delito y Victimización (National Strategy for the Prevention of Crime and Violence 2017-2027) ENPEVI Spanish acronym for Encuesta Nacional de Percepción de Seguridad Pública y Victimización (National Survey on Perceptions of Public Security and Victimization) EQ Evaluation Question ET Evaluation Team FGR Spanish acronym for Fiscalía General de la República (Office of Attorney General of the Republic) FLACSO Spanish acronym for Facultad Latinoamericana de Ciencias Sociales (Latin American School of Social Sciences) FMLN Spanish acronym for Frente Farabundo Martí para la Liberación Nacional (Farabundo Martí National Liberation Front) FUNDAUNGO Spanish acronym for Fundación Dr. Guillermo Manuel Ungo (Dr. Guillermo Manuel Ungo Foundation) FUNDE Spanish acronym for Fundación Nacional para el Desarrollo (National Development Foundation) vi FUSADES Spanish acronym for Fundación Salvadoreña para el Desarrollo Económico y Social (Salvadoran Foundation for Economic and Social Development) GDP Gross Domestic Product GOES Government of El Salvador GOH Government of Honduras GOG Government of Guatemala IDHUCA Spanish acronym for Instituto de Derechos Humanos de la Universidad Centroamericana (Human Rights Institute of University of Central America) IML Spanish acronym for Instituto de Medicina Legal (Institute of Forensic Medicine) INE Spanish acronym for Instituto Nacional de Estadísticas (National Institute of Statistics) INEGI Spanish acronym for Instituto Nacional de Estadística, Geografía e Informática (National Institute of Statistics, Geography, and Informatics) InfoSegura Regional Citizen Security Project IOM International Organization for Migration IPCESC Spanish acronym for Índice de Priorización de Centros Educativo en Materia de Seguridad Ciudadana (Prioritization Index on Educational Centers related to Citizen Security Issues) IUDPAS Spanish acronym for Instituto Universitario de Democracia, Paz y Seguridad (University Institute of Democracy, Peace, and Security) IUDOP Spanish acronym for Instituto Universitario de Opinión Pública (University Institute for Public Opinion Research) KII Key Informant Interview LAV-UERJ Portuguese acronym for Laboratório de Análise da Violência-Universidade do Estado do Rio de Janeiro (State University of Rio de Janeiro Laboratory of Violence Analysis) LEIV Violence-Free Life for Women M&E Monitoring and Evaluation MINED Spanish acronym for Ministerio de Educación (Ministry of Education) MINGOB Spanish acronym for Ministerio de Gobierno (Governance Ministry) MJSP Spanish acronym for Ministerio de Justicia y Seguridad Pública (Ministry of Justice and Public Safety) MTI Spanish acronym for Mesa Técnica Inter-institucional (Interinstitutional Technical Committee) NGO Non-Governmental Organization ODS Spanish acronym for Objetivos de Desarrollo Sostenible (Sustainable Development Objectives) OMSC Spanish acronym for Observatorio Municipal de Convivencia y Seguridad Ciudadana (Municipal Observatory for Citizen Coexistence and Security) PESS Spanish acronym for Plan El Salvador Seguro (Plan for Secure El Salvador) PNC Spanish acronym for Policía Nacional Civil (National Civilian Police) PNUD Spanish acronym for Programa de las Naciones Unidas para el Desarrollo (United Nations Development Programme) vii PREPAZ Spanish acronym for Ministerio de Justicia y Seguridad Pública (Ministry of Justice and Public Security) RMSE Root Mean Average Squared Error SDG Sustainable Development Goal SEPOL Spanish acronym for Sistema de Estadísticas Policiales (Police Statistics System) SETEPLAN Spanish acronym for Secretaría Técnica y de Planificación (Technical and Planning Secretariat) SOW Statement of Work STU Sub-Technical Unit UAIP Spanish acronym for Unidad de Acceso a la Información Pública (Access to Public Information Unit) UCA Spanish acronym for Universidad Centroamericana José Simeón Cañas (Central American University José Simeón Cañas) UGAGLO Spanish acronym for Unidad de Gestión y Apoyo a los Gobiernos Locales (Unit for Administration and Support to Local Governments) UN United Nations UNAH Spanish acronym for Universidad Nacional Autónoma de Honduras (National Autonomous University of Honduras) UNDP United Nations Development Programme UNFPA United Nations Fund for Population Activities UNHCR United Nations High Commissioner for Refugees UNODC United Nations Office on Drugs and Crime UPCV Spanish acronym for Unidad para la Prevención Comunitaria de la Violencia (Unit for Community Prevention of Violence) U.S. United States US$ United States Dollars USAID United States Agency for International Development USAID/CAM United States Agency for International Development Central America and Mexico UTECI Spanish acronym for Unidad Técnica de Coordinación Interinstitucional (Technical Unit for Inter-Institutional Coordination) viii EXECUTIVE SUMMARY EVALUATION PURPOSE AND EVALUATION QUESTIONS The purpose of this performance evaluation of the United States Agency for International Development (USAID) Regional Citizen Security Project (InfoSegura) is to document project achievements and challenges to date, inform adjustments for the remainder of project implementation, and help with appropriate planning for future citizen security work. InfoSegura, implemented by the United Nations Development Programme, has two components: 1) increase evidence-based policymaking and 2) increase regional collaboration in citizen security, knowledge management, and networking. The project, funded at (US$)21 million, runs from May 2014 to May 2020. This evaluation covers the period May 30, 2014 through September 30, 2018. The evaluation seeks to answer the following three broad evaluation questions (EQs): 1. To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? 2. To what extent have Central American governments used crime and violence data— produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to effective citizen security-related policy decisions? 3. To what extent has the creation of partnerships and transparency (Knowledge Network on Citizen Security, or CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy? PROJECT BACKGROUND InfoSegura focuses on the Northern Triangle of Central America (El Salvador, Honduras, and Guatemala) with smaller interventions in Belize, the Dominican Republic, and Costa Rica1 . Central America is the least developed sub-region in the hemisphere in socioeconomic terms with relatively high levels of both poverty and inequality. In recent times, the three countries have experienced some of the world’s highest rates of homicide and violence, with crimes against women, including femicide, increasing at a faster pace than homicide rates. The high crime rates, and resulting high levels of emigration, constitute key development challenges. The recent surge in migration to the United States from the Northern Triangle, and the inability to find solutions, are primary manifestations of these challenges. DESIGN, METHODS, AND LIMITATIONS The information for this evaluation comes from three primary sources: 1) document review; 2) key-informant interviews (KIIs) of project stakeholders; and 3) analysis of secondary data. KIIs were open-ended and semi-structured around the three EQs. KIIs included 30 persons in El Salvador, 17 in Honduras, and 19 in Guatemala. The full evaluation team (ET) participated in all interviews, with each team member taking handwritten notes. 1 For the purposes of this evaluation only the three countries of the Northern Triangle of Central America (El Salvador, Guatemala, and Honduras) were evaluated. ix Methodological limitations to the evaluation included: 1) absence of a counterfactual to assess attribution; 2) findings based largely on subjective perceptions of key informants interviewed using qualitative data collection methods; 3) difficulty assessing the sustainability of observed results; and 4) the potential for respondent and/or interviewer bias. x Table A1: Summary of Findings, Conclusions, and Recommendations EL SALVADOR Findings Conclusions EQ 1: To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? InfoSegura has helped to consolidate data in the office of Management of Research and Analysis (DIA) of the Ministry of Justice and Public Safety (MJSP). From the USAID perspective, harmonizing crime data across different sources and making it available to users are key issues addressed by the project. The DIA website is a notable achievement in this direction. 1. El Salvador is where InfoSegura has had the greatest impact in supporting the development of evidence-based public policies for citizen safety at a national level. 2. There is a need to revise the methodology and conduct of the Crime Victimization Survey to ensure that it is based on a properly constructed sampling frame to avoid coverage errors. The survey report, moreover, can be improved to ensure it contains relevant findings and sufficient information for users to assess the data quality. Other improvements can be made to the structure and writing of the report. 3. Much work remains to achieve a system of crime statistics containing a minimal set of homologated crime and violence data. To date, all efforts have been devoted to homicide counts. At the same time, inconsistencies exist in homicide data reported by the PNC crime analyzer and the DIA website and by the crime analyzer and the PNC’s Access to Public Information Unit (UAIP). • The development of evidence-based public policies requires an understanding of many citizen safety issues, including their measurement. There is almost a total absence of specialized input from a criminological perspective as well as from the field of official crime statistics. A criminological perspective should not be seen as divorced from the UNDP’s paradigm for citizen security. It provides the theoretical background to a view of citizen security as a democratic citizen order that eliminates the threat of violence in the population and allows for peaceful coexistence. 4. The fact that most of the institutions touched by the InfoSegura project acknowledge the sustainability of the actions that have been implemented is encouraging. There is no doubt that InfoSegura-supported initiatives have addressed relevant aspects of what is and will remain the major public policy problem in El Salvador. InfoSegura has supported the development of an advanced crime analyzer in the National Civilian Police (PNC). InfoSegura has supported development of both Municipal and School Prioritization Indices. InfoSegura has supported the harmonization of PNC geographical units with population census geographical segments. InfoSegura has supported data-systems integration and geocoding. The PNC is implementing a system that integrates geocoded crime data produced at the level of delegations, sub-delegations, and police posts into the institutional crime statistics at the Center of Police Operations. InfoSegura has helped to integrate a permanent monitoring system into the Plan for a Secure El Salvador (PESS) EQ1.1: To what extent has InfoSegura developed the capacity within the sub-technical units (STUs) (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? InfoSegura has helped to consolidate data into the DIA. EQ1.2: To what extent has InfoSegura developed the capacity of source institutions [such as the PNC, Attorney General’s Office (FGR), and Forensic Medicine Institute (IML) in El Salvador] to ensure quality of data and improve information sharing between source institutions and STUs? InfoSegura has helped structure existing pre-InfoSegura protocols between the PNC, the IML, and the FGR. xi Findings Conclusions InfoSegura “Technical #16. supported the Technical Secretariat of Planning’s (SETEPLAN) Committees” for Sustainable Development Goals (SDGs) #5 and EQ1.3: What is the potential that the capacity developed within STUs and source institutions increases or decreases after the completion of InfoSegura? InfoSegura helped strengthen capacities, systems, and tools to continue serving policymakers in the future, independent of political transitions. EQ2: To what extent have Central American governments used (as planned) crime and violence data produced by InfoSegura, and by other institutions that have received technical assistance under InfoSegura, as input to effective citizen security-related policy decisions? InfoSegura has made a significant contribution to the formulation of citizen security-related policymaking in the diverse forms of support it has provided to El Salvador’s national crime and violence prevention plan and PESS and its monitoring system and in developing the index to prioritize municipalities to be included at each of the three phases of PESS. 1. 2. InfoSegura’s support has provided key inputs for policymaking in the citizen security sector through: 1) its contribution to PESS; 2) its provision of municipal prioritization indices that permit targeted interventions and policies tailored to the needs of these areas; 3) its robust design of a PESS monitoring system, which allows policy monitoring and evaluation (M&E); and 4) its monitoring report, which provides reliable and concrete data on progress and ongoing challenges. InfoSegura has generated an important and groundbreaking body of data and analysis regarding violence towards women in El Salvador that is informing policymaking in the area and already fills a legislative requirement. The project’s support to developing a monitoring system for Goal #16 of the SDGs will contribute to shaping policymaking in order to achieve this objective. A number of factors foster or impede the process. The former include the existing policy frameworks and legislation for which data is key. The PESS framework, in particular, has made it possible to develop and tailor data to Reflecting InfoSegura’s commitment to make gender a centerpiece, the project generated an important and groundbreaking body of data and analysis regarding violence towards women in El Salvador that is informing policymaking in the area and responding to the legislative mandate to develop and implement an information system on violence against women. InfoSegura’s activities in support of establishing a monitoring process of the SDG #16 agenda will permit SETEPLAN to generate key inputs to security sector policymaking focused on achieving this goal. InfoSegura supported the development of a methodology for a Prioritization Index on Educational Centers Related to Citizen Security Issues (IPCESC), a range of specific policy needs and requirements. A factor that impedes the which provides data on variables on security risks to schools and has provided greater use of data for policymaking, and which is also evident in the other the Ministry of Education (MINED) and the MJSP timely and concrete data for two countries, is the lack of analytical capacity in many STUs and source targeting priority at-risk schools and developing relevant policy and action institutions for this data. plans. EQ2.1: What are the internal and external factors that foster or impede the governments to use the data? Factors that serve to foster the use of data and legislation for which data is key. are existing policy frameworks A factor that impedes the greater use of data for policymaking is the lack of xii Findings Conclusions analytical capacity in many STUs and source institutions for this data. Perennial issues that threaten the ongoing use of data for public policymaking are staff rotation/turnover and changes in public policies. An additional impediment in the use of data for evidence-based policymaking is the weakness of the General Directorate of Statistics and Census (DIGESTYC). There is a lack of clear mechanisms or information to policymakers. guidelines for the channeling of HONDURAS Findings Conclusions EQ 1: To what extent has InfoSegura collection, and recording in the region? contributed to change the level of quality and timeliness of crime and violence data definition, InfoSegura has strengthened technical roundtables on violent deaths, domestic abuse/family violence, sexual abuse and crime, and victimization and violence prevention at the Technical Unit for Inter-Institutional Coordination (UTECI) and at 30 operational municipal observatories. 1. 2. Honduras is where InfoSegura has had the greatest impact in supporting the development of evidence-based public policies for citizen safety at a local level. The project has focused on developing and strengthening local capabilities to produce homologated crime data on 11 types of incidents. Some efforts to use crime data to support policy-oriented crime research by CEASCI are taking place. A major weakness according to the ET’s perspective has to do with the lack of trained professionals in fields related with the sociology of crime and criminology. The ET recognizes that the InfoSegura project was not set-up based on a model of criminology. A diploma in criminalistics and criminology was provided through UNAH for 120 officials of the Municipal Observatories for Citizen Coexistence and Security (OMSC) in 2016 and these officials could provide support to InfoSegura. InfoSegura has strengthened official statistics on 11 types of violence. EQ1.1: To what extent has InfoSegura developed the capacity within STUs (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? InfoSegura-funded staff have been incorporated source institutions. into positions within STUs and EQ1.2: To what extent has InfoSegura developed the capacity of source institutions (such as the PNC, FGR, and IML in El Salvador) to ensure quality of data and improve information sharing between source institutions and STUs? UTECI, Center for the Study and Analysis of Coexistence and Citizen Security (CEASCI), and Unit for Administration and Support to Local Governments (UGAGLO) use crime data to conduct research and make policy at both the national and municipal levels. InfoSegura in 2016. has enhanced data collection and the conciliation protocol of UTECI xiii Findings Conclusions EQ1.3: What is the potential that the capacity developed within STUs and the source institutions increases or decreases after the completion of InfoSegura? Institutional capacities of STUs, local police statistics system units, and municipal observatories are being consolidated. EQ 2: To what extent have Central American governments used (as planned) crime and violence data produced by InfoSegura, and by other institutions that have received technical assistance under InfoSegura, as input to effective citizen security-related policy decisions? InfoSegura has made a significant contribution to citizen security-related policymaking via development of the Municipal Prioritization Index. 1. InfoSegura has contributed to an effective use of data and information to inform evidence-based policymaking at the municipal level. InfoSegura’s support to the Municipal Prioritization Index has enabled the identification of 30 municipalities for the establishment of OMSCs. Under the leadership of the UTECI, and with InfoSegura’s support, local governments are using crime data to develop and monitor crime prevention initiatives. 2. The limited capacity for analysis in different state entities which could be making use of available data impedes the more effective use of data for policymaking in Honduras. Crime prevention initiatives are being developed and monitored at the local level. This is a significant achievement as it allows for the effective involvement and consequent empowerment of local actors in the identification of strategies for crime control and the implementation of promising social and situational crime prevention initiatives. EQ2.1: What are the internal and external factors that foster or impede the governments to use data? An intensely polarized political situation gave way to a complex election context in 2017 which brought major policy changes in the area of citizen security, shifting the policy orientation towards the local level. Some elements that impede the use of data for policymaking are linked to the lack of trained professionals in fields related with the sociology of crime and criminology. The limited capacity for analysis in relevant government entities impedes the use of data for policymaking. GUATEMALA Findings Conclusions EQ 1: To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? InfoSegura’s Municipal Prioritization Index was used to establish pilot observatories like the one in Chimaltenango Department with plans to expand the model in other departments. 1. InfoSegura has supported the development of evidence-based public policies for citizen safety at a regional-departmental level. The project has focused on developing and strengthening local Data from the InfoSegura-supported Crime Victimization Survey were used to xiv Findings Conclusions prepare five infographic reports on major findings, to be published on final approval. capabilities to produce homologated crime data on eight types of incidents. 2. Despite political issues beyond its control, the project has managed to continue to support the implementation of initiatives aimed at improving the quality and availability of crime data. EQ1.1: To what extent has InfoSegura developed the capacity within STUs (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? The Government of Guatemala (GOG) has taken full advantage of InfoSegura￾supported processes to build capacities for crime prevention in relevant STUs. InfoSegura has strengthened the PNC statistics unit by providing it computing equipment and training on statistics and data analysis. The statistics unit is undergoing a process to achieve data-quality certification. EQ1.2: To what extent has InfoSegura developed the capacity of source institutions to ensure quality of data and improve information sharing between source institutions and STUs? InfoSegura sponsored several training events on data production and data quality for diverse institutions—e.g., Inter-institutional Technical Committee, National Institute of Statistics (INE), and Unit for Community Prevention of Violence (UPCV). InfoSegura has supported the consolidation of the inter-institutional technical roundtable and the “Sub Mesa” of data validation based on the UTECI protocol, which will yield crime statistics (eight crimes detailed) to be transferred to the INE. InfoSegura has strengthened the PNC statistics unit through technical assistance. EQ1.3: What is the potential that the capacity developed within STUs and the source institutions increases or decreases after the completion of InfoSegura? The STUs at the Vice Ministry of Crime Prevention have enjoyed staff stability despite political issues. However, significant staff turnover within the executing units of the Vice Ministry - PNC and UPCV - may make it difficult to ensure that key processes are maintained. EQ 2: To what extent have Central American governments used (as planned) crime and violence data produced by InfoSegura, and by other institutions that have received technical assistance under InfoSegura, as input to effective citizen security-related policy decisions? InfoSegura has contributed to the formulation of citizen security-related policy decisions through its contributions to the National Strategy for the Prevention of Crime and Violence 2017-2027 (ENP), including developing the index to prioritize 1. InfoSegura has contributed to a qualitative leap in data and information use to inform evidence-based policymaking. xv Findings Conclusions municipalities to be targeted for interventions and identifying key locations for departmental crime observatories, with pilots in Chimaltenango and Guatemala City. 2. InfoSegura’s support to the Municipal Prioritization Index, which underpins strategic interventions in the ENP, has also been used to establish pilot observatories like the one in Chimaltenango Department and has been a key underpinning for the ENP. 3. The uncertain political situation in Guatemala, which potentially affects all state institutions, could pose an impediment to some of the processes put in place with InfoSegura’s assistance and the government’s use of data for policymaking in citizen security and violence prevention. 4. Despite these uncertainties, and delay in the publication of the ENPEVI survey, the InfoSegura team continues to support the ongoing discussion of the use of these knowledge products for policy formulation. 5. The insufficient capacity for analysis in different state entities which could be making use of available data impedes the more effective use of data for policymaking. InfoSegura provided key support for the National Survey on Perceptions of Public Security and Victimization (ENPEVI), the first national survey of its kind in the NTCA, developed with both the Governance Ministry (MINGOB) and the INE. This survey is a key component of the ENP and will provide crucial data to underpin policy decisions. InfoSegura contributed to the development of a Model for Departmental Observatories on Citizen Security and Violence Prevention in Guatemala, permitting the launching of a pilot observatory in Chimaltenango and Guatemala City. This institutional model will serve to provide objective, reliable, up-to-date data to inform departmental plans and strategies to more effectively confront crime and violence and their causal factors. EQ2.1: What are the internal and external factors that foster or impede the governments to use data? The political situation in Guatemala has created an uncertain national and institutional environment in general, but a positive note for InfoSegura in the context of these institutional uncertainties is that the technical staff of the Vice Ministry seems to enjoy a relatively high degree of stability. The insufficient capacity for data analysis in different state entities was cited by government staff as an impediment to using data for policymaking. EL SALVADOR, HONDURAS, AND GUATEMALA Findings Conclusions EQ3: To what extent has the creation of partnerships and transparency (CONOSE) between government and civil society led to changes in national and regional collaboration with regard to the design of citizen security policy? InfoSegura has made important strides in overcoming the historical lack of trust between government and civil society, which has, in specific instances, led to collaborative relationships between CONOSE institutions, civil society, and government. 1. The CONOSE network has made important strides in overcoming the historical distrust between government and civil society regarding possibilities for collaborative efforts in joint problem￾solving and has undertaken efforts to both: 1) build stronger inter￾institutional relationships at a regional level; and 2) strengthen civil society and academia’s capacity for research and training. Nonetheless, important challenges still exist with respect to the InfoSegura has undertaken a range of diverse initiatives in the areas of regional fora, training activities, conferences, and workshops, reflecting a recognition of the urgent need to strengthen training and research capacities in the region. xvi Findings Conclusions EQ3.1: What are the internal and external factors that foster or impede collaboration among the members of CONOSE? construction of a joint research agenda between academia and government policymakers in the citizen security sector. A variety of impediments continue to prevent CONOSE from fully achieving its principal goals: 1) missing to date is a joint research agenda between state and civil society; 2) the absence of research within the framework of criminology and the absence of training opportunities in criminological approaches; 3) the CONOSE network structure and lack of legal status compromise access to broader funding opportunities, which are critical for chronically underfunded research institutes and universities. Recommendations General Recommendations • Training programs in the fields of official crime statistics, survey sampling, crime measurement scales, crime analysis for crime investigation and police operations, and evaluation of crime prevention policies and programs should be designed and implemented in the short term. • InfoSegura should strengthen analytical capacity for evidence-based policymaking within relevant institutions to ensure the data available is fully exploited. With an increased focus on multidimensional analyses of citizen security and development such as the study of homicide and poverty conducted by CEASCI in Honduras, the challenge of translating this material into effective, concrete policies is even greater. • Efforts to provide training on the development of databases that integrate crime data with data on social and economic issues, and the development of capacities to link data with public policy should be among the future priorities of InfoSegura. • Given the threats to the sustainability of results due to major electoral changes or ongoing political volatility, InfoSegura should seek to strengthen institutional relationships through formal agreements, maximizing institutional commitments to policy frameworks and information sharing to consolidate the institutional architecture and spaces established by the project. • InfoSegura should identify priority areas for consolidation of critical areas of work in the three countries, rather than launching diverse new activities, which may result in dispersion. Identifying the principal weaknesses that serve as obstacles for the effective realization of core goals could provide a road map for concentrating efforts. • As InfoSegura has generated significant data and analysis on gender and violence against women, it should further help develop a knowledge management systematization to identify lessons learned and good practices in mainstreaming of gender in information management for citizen security. Country-Specific Recommendations EQ1: El Salvador • There is a need to develop a crime and violence statistical system in El Salvador under the leadership of a specialized institution whose responsibilities include all the aspects involved in policy-oriented crime research. This institution could take on the role of temporary producer of crime statistics until such a stage when the system is fully developed and transferred to DIGESTYC. • Since the DIGESTYC survey is the first source of official data on the topic, the unit record data should be made available to the public to allow its use by xvii Recommendations researchers and public policy analysts. The same applies to the data files of the Violence Against Women Survey. • The lack of structured training on the production of crime statistics, its use for crime analysis to support police operations and crime investigations, and its further use to develop tools to support evidence-based citizen safety public policies is a major issue requiring InfoSegura’s attention. • The need to provide high quality training on the use of crime data for different purposes requires the development of a specialized occupational structure in the fields of official crime statistics, crime analysis, and crime research. The same institution mentioned above could assume leadership working in tandem with other institutions, such as the National Academy for Public Safety and the Penitentiary Academy, the Attorney’s School within the FGR, other specialized training organizations, and public and private tertiary education institutions. • More emphasis needs to be placed on the use of data found in the DIA web portal, which represents a notable advancement in access to updated information on the critically important, yet historically neglected, issues of violence against women and human trafficking. EQ1: Honduras • There is a need to provide high quality training on the use of crime data for different purposes, which requires the development of a specialized occupational structure in the fields of official crime statistics, crime analysis, and crime research. The University Institute of Democracy, Peace, and Security (IUDPAS) and/or National Autonomous University of Honduras (UNAH) could assume the leadership in this initiative. EQ1: Guatemala In light of the ongoing political uncertainties in the country, the UNDP in its role as an inter-institutional and impartial articulator should continue to seek to consolidate InfoSegura’s achievements thus far and strengthen the important institutional alliances surrounding data generation. The InfoSegura team should continue building up its technical capacities in the process of generating statistical knowledge products, which should provide the processes implemented by InfoSegura with some sustainability. EQ2: Honduras • InfoSegura should identify the areas in the institutional architecture that can most strengthen and further consolidate the work of UTECI and the other STUs, including working with CEASCI to use the multidimensional study it is completing as an input to evidence-based policymaking. EQ2: Guatemala • There is a need to strengthen analytical capacity within relevant public institutions for policymaking in law enforcement and violence strategies. InfoSegura should seek to strengthen the Departmental Observatories, providing key technical and other support to the new pilot observatory in Chimaltenango, and to support the Certification of Statistical Quality process for the PNC. EQ3 • CONOSE needs to take steps to ensure that research activities increasingly respond to a common agenda between government officials and academia, such as by holding a forum on establishing this agenda and exploring the current gaps that exist between government needs for evidence-based studies on the effectiveness of crime prevention initiatives/strategies versus the tendency in academia to pursue more theoretical concerns. • InfoSegura should examine the possibilities of collaborating with universities and experts in the hemisphere to expand curricula to include such courses as well as inviting select experts to conduct seminars and courses in this area. • InfoSegura needs to explore mechanisms by which CONOSE could receive outside funding given the obstacle created by CONOSE’s loose network structure and lack of legal status as an institution. 1 1.0 EVALUATION PURPOSE AND QUESTIONS 1.1 EVALUATION PURPOSE This evaluation report presents primary findings, conclusions, and recommendations from the performance evaluation of the United States Agency for International Development in Central America and Mexico (USAID/CAM) Regional Citizen Security Project (InfoSegura). The purpose of the evaluation is to document project achievements and challenges to date, to inform USAID of any adjustments for the remainder of project implementation, and to help with appropriate planning for future citizen security work. In addition, the evaluation will provide empirical evidence on management issues and support learning and continuous improvement in USAID’s regional citizen security work for current and future activities. InfoSegura’s time period is from May 2014 to May 2020, with a total investment of United States dollars (US$)21 million. This evaluation covers the period May 30, 2014 through September 30, 2018. The principal audiences for this evaluation are the United Nations Development Programme (UNDP), InfoSegura’s implementing partner, and USAID/CAM, including its Democracy and Governance Office and Regional Program Office. Both UNDP and USAID/CAM will use the evaluation’s results to make necessary adjustments to the project over its remaining life. USAID/CAM will further use the evaluation’s results to determine areas and approaches for future regional citizen security activities based on USAID/CAM’s strategic goal and development objectives. Other USAID offices with interest in the evaluation results are the Democracy and Governance Offices within USAID/Guatemala and USAID/Honduras; the Bureau for Latin America and the Caribbean (particularly the team that coordinates the Central America Regional Security Initiative); and the Bureau for Democracy, Conflict, and Humanitarian Assistance. In addition, evaluation results will be used for reporting purposes to stakeholders, including central and local governments in the Northern Triangle, other United States (U.S.) Government agencies, other bilateral and multilateral donors working on citizen security issues, civil society organizations, and academia. 1.2 EVALUATION QUESTIONS The evaluation seeks to answer three broad evaluation questions (EQs) that lie along a four-point axis of data production, data analysis, data usage, and data dissemination. The EQs, along with related sub-questions, are listed below. 1. To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? 1.1 To what extent has InfoSegura developed the capacity within the sub-technical units (STUs), located within each country’s respective Security Ministry, to collect, record, and use crime and violence data for analysis? 1.2 To what extent has InfoSegura developed the capacity of source institutions (such as the National Civilian Police, the Attorney General’s Office, and the Forensic Medicine 2 Institute in El Salvador) to ensure quality of data and improve information sharing between source institutions and STUs? 1.3 What is the potential that the capacity developed within STUs and the source institutions increases or decreases after the completion of InfoSegura? 2. To what extent have Central American governments used (as planned) crime and violence data—produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions? 2.1 What are the internal and external factors that foster or impede the governments to use the data? 3. To what extent has the creation of partnerships and transparency (Knowledge Network on Citizen Security, or CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy? 3.1 What are the internal and external factors that foster or impede the collaboration among the members of CONOSE? 2.0 PROJECT BACKGROUND 2.1 PROBLEM AND PROJECT CONTEXT Central America is the least developed sub-region in the hemisphere in socioeconomic terms with poverty rates higher than elsewhere in Latin America. The region also suffers from vast income inequality. According to 2011 World Bank data, 20 percent of the wealthiest segments of the population account for more than half of overall income, 2 while chronic undernutrition is also a major challenge. Given Central America’s proximity to the U.S., its problems directly affect U.S. interests. The countries of the Northern Triangle of Central America —El Salvador, Honduras, and Guatemala—have had a particularly rocky transition to democracy following the demise of military regimes three decades ago. In more recent years, the three countries have experienced some of the world’s highest rates of homicide and violence. El Salvador has the world’s highest homicide rate of youth up to 19 years, followed closely by Guatemala3 . El Salvador registered 6,640 killings in 2015, with monthly totals topping those seen during the country’s civil war, which ended in 1992. Disappearances—largely untracked and invisible—have also been on the rise. Further, aggression against journalists and human rights defenders has increased in the region, notably so in Honduras and Guatemala, while small businesses continue to close due to extortion or violence, leaving thousands without jobs or economic livelihoods.4 This violence, with prevailing high rates of impunity, is generated by gangs, drug trafficking, and trafficking in persons, although studies5 have shown that the crimes have multiple drivers. The 2 http://data.worldbank.org/indicator/NY.GDP.PCAP.CD?order=wbapi_data_value_2012+wbapi_data_value+wbapi _data_value-last&sort=asc 3 INSIGHTCRIME. “Ten Countries with Highest Child Homicide Rates all in LatAm: Report. June 6, 2017. 4 USAID/CAM Regional Development Cooperation Strategy, 2015-2019, p. 6. 5 Desk Study. 3 Crimes vary in type, but notably include high rates of femicide—decreasing at a slower pace than homicide rates—and other crimes against women. Only a fraction of total crimes against women are reported because of fear of retaliation from perpetrators, of economic dependence on perpetrators, and of lack of protection services. In some countries, the legal procedures are complicated and often require women to make long and repeated trips to report their cases. The high crime rates and resulting high levels of emigration constitute key development challenges. The recent surge in migration to the U.S. from Honduras, El Salvador, and Guatemala is but one result of these challenges and the inability, to date, to find solutions. It is in this general context that USAID/CAM funded the United Nations Development Programme Regional Initiative, InfoSegura, in 2014 to address citizen security issues in each of the three Northern Triangle Countries. Each country, however, offers a different context, and that country context is always changing. For more information on each of the three countries’ contexts, please go to Annex II. 2.2 THEORY OF CHANGE HYPOTHESIS AND CAUSAL LOGIC InfoSegura’s objectives are to: 1) strengthen public sector information management that would lead to evidence-based citizen security policymaking; 2) promote intersectional dialogue between government and civil society in the design of citizen security public policy; and 3) enhance regional collaboration, knowledge management, and networking on citizen security. Expressed differently, InfoSegura aims to provide single-source reliable crime data in each of the three countries that would allow for comparisons from one year to the next in each country as well as for cross￾country comparisons. This, in turn, would allow the development of sound public policy at all levels and allow civil society to be producers and consumers of reliable information. (The theory of change hypothesis and results framework for InfoSegura are presented in Annex IV.) 3.0 EVALUATION METHODS AND LIMITATIONS 3.1 EVALUATION METHODS InfoSegura is built around four sequential activities: 1) data production, 2) data analysis, 3) data use, and 4) data dissemination. This four-phase activity model remained uppermost in the minds of the ET and guided it throughout the evaluation. With that in mind, the InfoSegura performance evaluation used a mixed-methods approach to data collection that included, principally, a document review and KIIs. Key data sources for the evaluation are listed in Table 1. 4 Table 1: Evaluation Data Sources Data Sources USAID in El Salvador, Honduras, and Guatemala UNDP in El Salvador, Honduras, and Guatemala Municipal Entities in El Salvador, Honduras, and Guatemala CONOSE in El Salvador, Honduras, and Guatemala Ministry of Security, Sub-Technical Unit in El Salvador, Honduras, and Guatemala National Civilian Police in El Salvador, Honduras, and Guatemala Attorney General’s Office in El Salvador, Honduras, and Guatemala Document Review The ET reviewed numerous documents, which are listed in the bibliography in Annex VI. Included among those documents are performance indicators prepared by UNDP corresponding to the project’s theory of change hypothesis and results framework in Annex IV. The ET also conducted a contextual literature research and review, which informed the regional and country background sections above and in Annex II. Key Informant Interviews KIIs are semi-structured, open-ended interviews with members of key project stakeholder groups and were almost always conducted in the offices of interviewees in the three Northern Triangle countries. The interviews posed questions that related, directly or indirectly, to the three EQs. Given the range of interviewees, the questions posed necessarily differed in accordance with interviewees’ interest in the project and their firsthand knowledge of relevant information. The entire ET participated in all of the interviews and took handwritten notes of responses. The responses were later charted on a matrix prepared and then analyzed for frequency of relevant themes corresponding to one or more of the EQs. KIIs included 30 persons in El Salvador, 17 in Honduras, and 19 in Guatemala. (See Annex III for the data collection instruments and Annex V for the evaluation matrix matching the EQs with data sources, data collection methods, and data analysis methods.) 3.2 EVALUATION LIMITATIONS The evaluation methodology involved the following methodological limitations: • Absence of a counterfactual to assess attribution. In some measure, this limitation is addressed through triangulation from multiple sources, i.e., documents and KIIs. The ET used a modified contribution analysis to address this limitation.6 • Evaluation findings are based largely on subjective perceptions of key informants interviewed using qualitative data collection methods. The ET was not able to verify all perceptions and anecdotes cited by key informants. The evaluation design sought to compensate for this 6 Contribution analysis pretends to offer enhanced rigor to apply the principles of attribution to situations in which the program is one of several potential contributory causes and where it is not possible to measure attribution directly via creation of a scientifically-valid counterfactual. Thus, in lieu of assessing “attribution,” contribution analysis assesses the “contribution” of program activities to observed outcomes. The theory of contribution analysis is not well developed and is applied here in only a general way. 5 limitation through triangulation achieved by generating data from multiple sources, including project reports, knowledge products, analysis tools, public policies, and progress reports. • Difficulty assessing the sustainability of observed results. Since InfoSegura has been extended, there is nothing “final” to assess for sustainability. Assessments were made on the basis of work completed at the time of the evaluation. • The potential for respondent and/or interview bias. Notwithstanding best efforts to avoid or mitigate potential biases existing among key informants, they remain a risk, particularly with regard to qualitative data collection methods, which rely heavily on respondent perceptions and interviewer interpretation. The evaluation design sought to compensate for this limitation through the application of best practices in data collection, cross-team validation, and systemic data analysis that synthesized multiple data sources using well￾established and credible data analysis methods. 4.0 FINDINGS, CONCLUSIONS, AND RECOMMENDATIONS This section presents the evaluation’s primary findings and conclusions drawing on information obtained during the document review and KIIs. The findings and conclusions for EQ1 and EQ2 are presented by country, while those for EQ3 are presented at a regional level. 4.1 EQ1 FINDINGS 4.1.1 EQ1: To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? EL SALVADOR InfoSegura supported the development of an advanced crime analyzer in the Policía Nacional Civil (National Civilian Police, or PNC). The advanced crime analyzer7 is a database system that enables users to submit queries on crimes reported to the PNC, including the number of violent and property crimes, alleged offenders arrested, victims, numbers of traffic accidents, and sexual offenses. The number of reported incidents can be obtained for the following types of offenses: extortion, homicide, manslaughter, theft, motor vehicle theft, transport theft, injuries, robbery, motor vehicle robbery, kidnapping, and rape. Statistics are available for each year from 2015 through 2018 and can be obtained for a wide variety of cross￾classifications. The PNC's analyzer and transparency site has also been supplemented by the MJS/DIA's own transparency webpage, with an advanced analyzer. So, there are essentially two platforms. However, the DIA's is much broader than the PNC's as it provides access to data on homologated homicide counts, violence against women, trafficking in human beings, prioritization tools such as the MPI and the School Prioritization Index, and the PESS Monitoring Plan8 . 7 The advanced crime analyzer is accessible through the following link: http://transparencia.pnc.gob.sv/ 8 The DIA data can be accessed through the following link: http://www.seguridad.gob.sv/dia/ 6 In addition, by clicking on the “offenses” label in the PNC’s transparency website, users gain access to graphics showing the number of recorded incidents in 2018 classified by month up to September, area (rural/urban), weapon involved, and province (department). At its current stage of development, the crime analyzer is a tool that enables users to obtain tables with counts along with some basic graphs to visualize trends. Data can be exported into Excel comma-delimited files. Several weaknesses were identified that impair a more efficient use of the tool. First, the site does not include documentation on crime definitions, counting rules, and processes to record victims and/or offenders. Second, it does not provide specific data for the offense of femicide. It could be argued that at the time an incident of violent death of women is recorded, the PNC does not have sufficient information to determine whether the death was a femicide, and that the FGR later determines whether the death fits the judicial definition. However, it would be feasible for the PNC to provide data for a category such as “alleged femicide.” Quality seems to be another issue with the data produced through the crime analyzer. The ET made a data request for the number of homicide and extortion victims. 9 The results of this inquiry, found in Table 2, show the number of victims obtained from the crime analyzer and the PNC’s Unidad de Acceso a la Información Pública (Unit of Access to Public Information, or UAIP). In the case of monthly homicides from 2015 through 2017, the counts obtained from the crime analyzer have been consistently smaller than those provided by the UAIP, while in 2018, the crime analyzer count was larger than the UAIP count. The discrepancies between the crime analyzer and the UAIP are worse for extortion rates. With the exception of 2017, the counts obtained from the analyzer were larger than those reported by the UAIP, with the difference being largest in 2015. The Root Mean Average Squared Error (RMSE)10 for homicide rates ranged from a minimum of 0.7 in 2018 to a maximum of 1.9 in 2016. For extortion, the RMSE ranged from a minimum of 13.1 in 2017 to a maximum of 50.5 in 2015. Ideally, and provided that the original data source was the same for both the crime analyzer and the UAIP, the RMSE should have a value of zero. A comparison of the 2017 data from the repository of homologated homicides, 11 available from the Dirección de Investigación y Análisis (Research and Analysis Directorate, or DIA) website, revealed a further weakness of the PNC’s crime analyzer. DIA data indicates that a total of 3,962 homologated homicides occurred in 2017. 12 The PNC’s crime analyzer gives a total of 3,954 9 Reference PNC-UAIP-837-2018. Data was requested on the number of victims for the offenses of homicide and extortion, classified by province (department), municipality, month, and sex of victims over the period from January 2013 through September 2018. 10 The RMSE is used to measure accuracy of variables in situations when large errors or discrepancies are particularly undesirable. The smaller the RMSE the more accurate (or less biased) is a statistical figure. 11 The violent deaths homologation table, made up of representatives from the PNC, FGR, and IML, meets every second fortnight to compare the institutional statistics for the preceding month, and to update any changes that might have occurred in previous months. Annual homicide statistics are consolidated in January of the following year. (http://www.seguridad.gob.sv/dia/estadisticas-homologadas/actas-de-la-mesa-de-homologacion-de-las-muertes￾violentas/ ) 12 Consolidated Homicides 2017. (http://www.seguridad.gob.sv/dia/estadisticas-homologadas/actas-de-la-mesa-de￾homologacion-de-las-muertes-violentas/ ) 7 homicides, eight less than DIA. The statistics from the PNC’s UAIP indicate a total of 3,962 homicides, the same number of homicides as DIA. Table 2: Counts of Victims of Homicide and Extortion by Data Source, January 2015 – September 2018 Month Homicide Extortion 2015 2016 2017 2018 2015 2016 2017 2018 Numbers of Victims According to Crime Analyzer January 336 740 259 323 185 191 144 172 February 307 664 241 308 178 216 144 143 March 484 611 323 312 182 170 127 141 April 418 353 290 318 179 206 97 112 May 643 356 299 242 210 194 123 108 June 677 336 370 304 192 186 122 139 July 470 374 320 291 214 169 101 121 August 918 402 339 276 196 184 135 119 September 685 343 441 195 188 151 136 139 October 684 407 451 200 188 179 November 449 361 279 165 197 156 December 585 333 342 153 131 124 Total 6,656 5,280 3,954 2,569 2,242 2,183 1,588 1,194 Numbers of Victims According to PNC’s UAIP January 339 741 259 322 128 149 171 139 February 308 666 241 308 136 155 135 123 March 485 614 323 312 146 121 138 103 April 420 354 293 318 148 152 96 97 May 644 357 298 242 151 160 130 106 June 678 336 371 304 131 156 120 139 July 471 374 320 292 160 154 111 128 August 920 405 340 276 151 181 136 120 September 686 347 442 193 142 143 134 136 October 684 407 452 142 186 148 November 450 362 280 111 189 153 December 586 333 343 100 127 124 Total 6,671 5,296 3,962 2,567 1,646 1,873 1,596 1,091 Difference (Crime Analyzer minus PNC’s UAIP) January -3 -1 0 1 57 42 -27 33 February -1 -2 0 0 42 61 9 20 March -1 -3 0 0 36 49 -11 38 April -2 -1 -3 0 31 54 1 15 8 Month Homicide Extortion 2015 2016 2017 2018 2015 2016 2017 2018 May -1 -1 1 0 59 34 -7 2 June -1 0 -1 0 61 30 2 0 July -1 0 0 -1 54 15 -10 -7 August -2 -3 -1 0 45 3 -1 -1 September -1 -4 -1 2 46 8 2 3 October 0 0 -1 58 2 31 November -1 -1 -1 54 8 3 December -1 0 -1 53 4 0 Total -15 -16 -8 2 596 310 -8 103 RMSE 1.4 1.9 1.1 0.7 50.5 33.2 13.1 16.4 InfoSegura has supported development of both Municipal and School Prioritization Indices.13 The Municipal Prioritization Index is a tool that assists in the identification of the municipalities that were a priority for the citizen safety and coexistence public policy. It was developed as a municipal weighted average of nine citizen safety and three risk-factor indicators. The index was used to classify municipalities into five levels of citizen safety: very low, low, medium, high, and very high. The Government of El Salvador (GOES) used this classification to select the 50 municipalities targeted for intervention in the three stages of the Plan El Salvador Seguro (Plan for a Secure El Salvador or PESS).14 The School Prioritization Index identifies public schools for targeting of strategic and public policy interventions by the Ministerio de Educación (Ministry of Education, or MINED) and Ministerio de Justicia y Seguridad Pública (Ministry of Justice and Public Safety, or MJSP). The index was constructed using data from the observatory on El Salvador’s public schools and the school census. The index has been used within the PESS framework to classify public schools into five categories according to their level of insecurity.15 InfoSegura has supported the harmonization of PNC geographical units with population census geographical segments. DIA is in the process of applying the method for prioritization of municipalities to the prioritization of police sectors. The homologation of PNC geographical units with those used by the Dirección General de Estadística y Censos (General Directorate of Statistics and Census, or DIGESTYC) to conduct the census of population and housing is a first step towards achieving this objective. The purpose is to standardize the police sectors in such a way that they match the census sectors. Such standardized police units would enable DIA to conduct more detailed analyses aimed at identifying areas within municipalities to be targeted for police interventions and for the development of crime prevention strategies at a micro level. According to an interview with a senior official, shortage of competences and abilities 13 InfoSegura nd. Herramienta para priorización y focalización de políticas públicas e materia de seguridad ciudadana. Aplicación en el marco del Plan El Salvador Seguro. 14 InfoSegura nd. Índice de priorización de municipios. Herramientas para la priorización y focalización de políticas públicas en materia de seguridad ciudadana. 15 InfoSegura nd. Índice de priorización de centros educativos. Herramientas para la priorización y focalización de políticas públicas en materia de seguridad ciudadana. 9 of the technical officers to conduct more sophisticated types of analyses and interpretations of multilevel spatial crime data is one factor impeding the further development of this tool. InfoSegura has supported data-systems integration and geocoding. The PNC is implementing a system16 that integrates geocoded crime data produced at the level of delegations, sub-delegations, and police posts into institutional crime statistics kept at the Center of Police Operations. The lack of trained crime analysts is a major barrier to completing this integration, however, which is a key component in assisting the Consejo Nacional de Seguridad Ciudadana y Convivencia (National Council for Citizen Security and Coexistence, or CNSCC) to monitor PESS. InfoSegura has helped to integrate a permanent monitoring system (e.g., monitoring indicators and Municipal Prioritization Index) into PESS. InfoSegura provided CNSCC with technical assistance to develop a monitoring system for PESS. It facilitated the hiring of the Fundación Dr. Guillermo Manuel Ungo (Guillermo Manuel Ungo Foundation, or FUNDAUNGO) to develop a monitoring system for PESS, which contained a number of technical cards for each of the 80 indicators related to the central themes of PESS. Data for 10 out of the 80 indicators is collected through the Crime Victimization Survey. Several other indicators have PNC’s administrative data as their source.17 InfoSegura has further supported the development of methodological tools for prioritization and targeting that have been widely used by the CNSCC in the implementation of the strategies of PESS.18 HONDURAS InfoSegura has strengthened technical roundtables on violent deaths, domestic abuse/family violence, sexual abuse and crime, and victimization and violence prevention at the Unidad Técnica de Coordinación Interinstitucional (Technical Unit for Inter-Institutional Coordination, or UTECI) and at 30 operational municipal observatories. Since 2017, InfoSegura has supported the publication of comprehensive Observatorio Municipal de Convivencia y Seguridad Ciudadana (Municipal Observatory for Citizen Coexistence and Security, or OMSC) bulletins for an increasing number of municipalities. The number of published bulletins increased from 10 in 2017 to 27 in 2018. InfoSegura has strengthened official statistics on 11 types of violence. InfoSegura has provided strong support for the development of UTECI’s technical and physical capacities for the collection, processing, and dissemination of crime statistics for 11 types of violent incidents. Also, InfoSegura’s support has been crucial to enhance the capacities of the OMSCs to produce local crime data. GUATEMALA InfoSegura’s Municipal Prioritization Index is used to establish pilot observatories like the one in Chimaltenango Department. The Municipal Prioritization Index identified the department of Chimaltenango as one with a strong concentration of municipalities with higher levels and risk of violence. Chimaltenango was selected as the seat for the first departmental 16 With assistance of the Bureau of International Narcotics and Law Enforcement Affairs (INL). 17 CNSCC, nd., Sistema de Monitoreo, Plan El Salvador Seguro. San Salvador. 18 Refer to finding 1.1.1 of this report. 10 observatory of crime and citizen safety under the InfoSegura project in 2018. It is in the process of implementing the model developed in 2017.19 Initial steps have been taken to develop relationships with the departmental actors. It is minimally staffed, but it is strongly supported by the Unidad para la Prevención Comunitaria de la Violencia (Unit for Community Prevention of Violence, or UPCV) within the Third Vice Ministry for Governance and the Prevention of Crime and Violence. Data from the InfoSegura-supported Crime Victimization Survey were used to prepare five infographic reports on major findings, to be published on final approval. The survey was funded by the Ministerio de Gobernación (Governance Ministry, or MINGOB) and designed by the Instituto Nacional de Estadísticas (National Institute of Statistics, or INE), while data was collected by a private company and processed with the assistance of the Mexican Instituto Nacional de Estadística, Geografía e Informática (National Institute of Statistics, Geography, and Informatics, or INEGI). InfoSegura supported the survey through the United Nations Office on Drugs and Crimes’ (UNODC) Center of Excellence and with the publication of the findings. At the time of writing, the survey findings have not been published pending MINGOB approval. 4.1.1.1 EQ1.1: To what extent has InfoSegura developed the capacity within STUs (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? EL SALVADOR InfoSegura has helped to consolidate data in the DIA of the MJSP. In 2014, the Salvadoran government sought InfoSegura’s cooperation to establish a unit within the MJSP charged with the analysis of data to contribute to the elaboration of strategies and decision￾making. In 2015, InfoSegura provided the MJSP provincial (departmental) delegations and the DIA with computing hardware and software. In 2017, DIA launched a strategic information website that was designed to provide civil society, higher education and research institutions, think tanks, and the media, among others, with access to justice and public safety statistics and information. The website offers information and statistics on the following citizen safety relevant topics: (a) homologated homicide statistics; (b) violence against women; (c) trafficking in human beings; (d) tools for prioritizing and targeting citizen safety-related public policies and monitoring and evaluation (M&E) systems for PESS; (e) Objetivos de Desarrollo Sostenible (Sustainable Development Objectives, or ODS), and the Development Five-Year-Plan; (f) municipal observatories; and (g) an advanced analyzer for producing statistics on homicide, kidnapping, and returned migrants.20 21 The DIA website is well designed and easy to use. From the USAID perspective, harmonizing crime data across different sources and making it available to users are key issues to be addressed by the InfoSegura project. The DIA website is a notable achievement in this direction. 19 InfoSegura. 2017. Modelo “Observatorios departamentales de seguridad ciudadana y prevención de la violencia en Guatemala.” Guatemala. 20 https://www.infosegura.org/2018/09/28/la-direccion-de-informacion-y-analisis-lanza-su-portal-de-informacion￾estrategica/ 21 http://www.seguridad.gob.sv/dia/ 11 HONDURAS InfoSegura-funded staff have been incorporated into positions within STUs and source institutions. Most of the people trained under the InfoSegura project are part of the staff of UTECI, Consejo Nacional de Estudio y Análisis de Convivencia y Seguridad Ciudadana (Center for the Study and Analysis of Coexistence and Citizen Security, or CEASCI) and Unidad de Gestión y Apoyo a los Gobiernos Locales (Unit for Administration and Support to Local Governments, or UGAGLO). This seems to be the case in most of the OMSCs. GUATEMALA The Guatemalan government has taken full advantage of the InfoSegura-supported processes to build capacities for crime prevention in relevant STUs. Staff hired under the project, moreover, are being incorporated into the MINGOB administrative structure. 4.1.1.2 EQ1.2: To what extent has InfoSegura developed the capacity of source institutions [such as the PNC, Fiscalía General de la República (Office of Attorney General of the Republic, or FGR), and the Instituto de Medicina Legal (Forensic Medicine Institute, or IML) in El Salvador] to ensure quality of data and improve information sharing between source institutions and STUs? EL SALVADOR InfoSegura has helped articulate existing pre-InfoSegura protocols between the PNC, FGR, and IML. The MJSP, through DIA, coordinates the inter-institutional technical table on violent deaths that homologates homicide data.22 This technical task brings together representatives of the PNC, FGR, and IML who meet every second fortnight to compare their homicide statistics for the previous month and to update changes that have occurred in previous months. Annual homicide statistics are consolidated in January of the following year. InfoSegura has assisted with the systematization of the homologation process, which has been transferred to be implemented in Honduras and Guatemala. InfoSegura has strengthened existing capacities within DIGESTYC to carry out national victimization and violence against women surveys. In 2017, InfoSegura, through the United Nations Fund for Population Activities (UNFPA), facilitated funding to DIGESTYC to conduct data collection for a crime victimization survey also known as the Culture of Peace Survey. InfoSegura commissioned FUNDAUNGO for survey design, processing, and implementation. Afterwards, InfoSegura published a document with graphics summarizing the survey’s primary findings. The survey used a questionnaire adapted from the VIC-LACSI, developed by 21 countries with support by IDB, UNDP and UNODC Center of Excellence as the standard to collect crime survey data in Latin America. InfoSegura and UNFPA supported a second crime victimization survey in 2018, with DIGESTYC taking on full responsibility for its implementation. UNFPA assisted with the management of the funds allocated for the payment of survey field staff. Major findings were published in September 2018.23 This was DIGESTYC’s first experience implementing all aspects of the crime victimization survey. Perhaps as a result, a number of concerns were raised about the quality of the survey. To 22 The technical table on homicides has been in operation since 2005. 23 DIGESTYC, 2018, Informe de resultados. Encuesta de victimización y percepción de inseguridad / encuesta cultura de paz. Ciudad Delgado: Ministerio de Economía. 12 begin with, the sampling frame for both the 2017 and 2018 surveys consisted of census segments not included in the sampling frame for the annual National Household Survey conducted by DIGESTYC owing in part to the lack of an updated cartography within the segments selected as primary survey sampling units. Furthermore, the survey report did not include either standard errors or any other quality criteria for the survey estimates; omitted minimum estimates related to repeat victimization, total numbers of incidents, reporting of crimes classified by type of incident and victim characteristics, and incident and offender characteristics; and used offense descriptions that differed from the 2017 survey. In addition to the crime victimization survey, InfoSegura assisted DIGESTYC in the preparation of the draft report for the Violence Against Women Survey in 2017. DIGESTYC published an undated document containing an executive summary with the main survey findings. Unlike the report for the 2018 Culture of Peace survey, the report for the Violence Against Women Survey included accuracy measures for the main estimates.24 InfoSegura supported the Secretaría Técnica y de Planificación’s (Technical Secretariat of Planning, or SETEPLAN) “Technical Committees” for Sustainable Development Goals (SDGs) #5 and #16. InfoSegura assisted SETEPLAN with the development of a monitoring system and technical cards, the provision of validated data, support to define infographics, and the development of a tutorial to facilitate the use of the information by the actors involved in the ODS process. HONDURAS UTECI, CEASCI, and UGAGLO used crime data to conduct research and make policy at both the national and municipal levels. In collaboration with the Centro Nacional de Información del Sector Social (National Center for Information on the Social Sector, or CENISS), CEASCI used UTECI’s data to prepare the Multidimensional Homicide Analysis Research Bulletin (Bulletin No. 1) and an ongoing study on risk factors associated with homicide in municipalities. The study aimed at identifying factors associated with homicides at the municipal level. The findings provided some evidence of a relationship between community social functioning and homicide rates. The second stage of this research seeks to develop an index of erosion of sociability to be used as a tool to prioritize and target municipalities for interventions of a new public policy approach to coexistence and public safety. InfoSegura supported the publication of findings. UGAGLO makes use of UTECI and Municipal Observatory data to support municipal crime prevention initiatives. Using UTECI-systematized data produced by the municipal observatories, and a diagnostic prepared by CEASCI, UGAGLO assists local governments to identify priorities for preparation of the municipal plan for public safety. Local actors and the municipal government identify priority local public safety problems and the responses to them. The initiatives are incorporated into a local safety plan that includes strategies for crime dissuasion, crime control, and social and/or situational crime prevention. Then UGAGLO certifies the local plan and funds are allocated for its funding. InfoSegura assists municipal observatories with the publication of bulletins and studies. 24 Gerencia de Estadísticas de Género, DIGESTYC, nd., Encuesta nacional de violencia contra la mujer. El Salvador 2017. Resumen Ejecutivo. San Salvador: MINEC, pp. 51-58. 13 InfoSegura enhanced data collection and the conciliation protocol of UTECI in 2016. InfoSegura supported the process of development and publication of a comprehensive protocol for the conciliation of data on 11 types of incidents: violent deaths, traffic accident deaths, suicides, sexual offenses, family and domestic violence, robberies, thefts, extortion, kidnapping, and injuries. This protocol defines the participant institutions and their roles, the composition of the technical roundtables, criteria for their functioning, the indictors for each of the thematic tables, criteria to disaggregate these indicators, forms for collecting data on each indicator, and the procedures to conciliate inter-institutional information. InfoSegura has further assisted UTECI with the publication of national reports and municipal bulletins detailing crime incidences and prevention actions. GUATEMALA InfoSegura sponsored several training events on data production and data quality for diverse institutions—e.g., Mesa Técnica Inter-institucional (Inter-institutional Technical Committee, or MTI), INE, and UPCV. InfoSegura-supported training activities have developed and strengthened INE’s capacities for the processing and estimation of data from administrative sources, such as those produced by the PNC and the FGR. UPCV staff were trained in the development and implementation of the model of crime observatories. Training has been provided to staff hired for the Chimaltenango crime observatory. Also, training activities have been conducted with staff of the PNC statistics unit. InfoSegura has supported the consolidation of the inter-institutional technical roundtable and the “Sub Mesa” of data validation based on the UTECI protocol, which will yield crime statistics (eight crimes detailed) to be transferred to the INE. Since 2017, official homologated crime data is produced through implementation of institutional conciliation protocols similar to those of Honduras. A statistical compendium of data on coexistence and crime prevention was published with InfoSegura support in 2018. InfoSegura has strengthened the PNC statistics unit through technical assistance. It has provided the statistical unit at the PNC with computing equipment and training on statistics and data analysis. The statistical unit is undergoing a process of accreditation to achieve data￾quality certification (Sello de Calidad, or Seal of Quality). Support for the Seal of Quality was undertaken jointly by the Center for Excellence, MINGOB, INEGI, UNODC, UNDP, as per inter￾agency agreement between UNDP and UNODC. 4.1.1.3 EQ1.3: What is the potential that the capacity developed within STUs and source institutions increases or decreases after the completion of InfoSegura? EL SALVADOR InfoSegura helped to strengthen capacities, systems, and tools that can continue to serve policymakers in the future, independent of political transitions. All of the interviewees concurred in the sustainability of InfoSegura contributions. Everyone agreed on the value that strengthened capacities, systems, and tools have as inputs to public policy and program development. 14 HONDURAS Institutional capacities of STUs, local Sistema de Estadísticas Policiales (police statistics system, or SEPOL) units, and municipal observatories are being consolidated. The whole public safety approach adopted by the Honduras government and its strong relationship with UNDP, USAID, and other donors has enabled InfoSegura to provide pertinent and efficient support for the production of crime statistics at the local level. GUATEMALA The STUs at the Vice Ministry of Crime Prevention have enjoyed staff stability despite political turnover, and there are indications of ongoing commitments to maintaining key processes. The Government of Guatemala (GOG) has taken full advantage of the InfoSegura-supported processes to build capacities for crime prevention, and staff hired under the project is being incorporated into the MINGOB administrative structure despite a significant staff turnover within the executing units of the Vice Ministry - PNC and UPCV. 4.2 EQ1 CONCLUSIONS EL SALVADOR El Salvador is where InfoSegura has had the greatest impact in supporting the development of evidence-based public policies for citizen safety at the national level. The monitoring system for PESS uses data from the crime victimization survey as measures for 10 of its indicators and uses PNC data as the source for other indicators as well. The existence of the CNSCC has widened the scope for InfoSegura contributions in the use of data as evidence for policymaking. Despite the novelty that conducting crime victimization surveys in 2017 and 2018 represented for El Salvador, and the apparent benefits of using an international standard methodology to both develop the survey instrument and process the survey data, the findings of the 2018 survey, as reported by DIGESTYC, suggest a number of technical areas requiring attention. First, there is the issue of ensuring that the survey is based on a properly constructed sampling frame to avoid coverage errors. Second, the report itself can be improved to ensure that it contains relevant findings and enough information for users to assess the quality of the data. Third, there are many improvements that can be made to the structure and writing of the report. The survey report did not include standard errors or any other quality criteria for the survey estimates; omitted minimum estimates related to repeat victimization, total numbers of incidents, reporting of crimes classified by type of incident and victim characteristics, and incident and offender characteristics; and used offense descriptions that differed from the 2017 survey. Much work remains to put in place a system of crime statistics containing a minimal set of homologated crime and violence data. At the time of this evaluation, all efforts have been devoted to homicide counts. Nevertheless, inconsistencies between homicide data made available to the public through the PNC crime analyzer and the DIA website continue to exist. Also, inconsistencies between the data produced by the PNC UAIP and the data from the crime analyzer have been detected. The development of evidence-based public policies requires an understanding of many citizen safety issues, including their measurement. Thus far, there has been almost a total absence of specialized input from a criminological perspective as well as from the field of official crime statistics. InfoSegura approaches violence as a human development issue linked to poverty, 15 inequality, and exclusion. The IT understands that InfoSegura was not designed to include a criminology perspective. However, since the project is about data on violent crimes, and homicide remains a major problem for Salvadoran society, an explanation for this phenomenon based on one of the major criminological theories (i.e. cultural-social learning, disorganization￾control, or anomie-strain) would be valuable. At the individual level, there is no defined theoretical approach guiding the project (i.e. routine activities, lifestyle, place-based approaches, etc.). The manner in which the crime victim survey was conducted and reported is an example of the lack of a theoretical framework. The fact that most of the institutions touched by the InfoSegura project acknowledge the sustainability of the actions that have been implemented is encouraging. There is no doubt that InfoSegura-supported initiatives have addressed relevant aspects of what is and will remain the major public policy problem for El Salvador. The sustainability of many of these initiatives will depend on political will. HONDURAS Honduras is where InfoSegura has had the greatest impact in supporting the development of evidence-based public policies for citizen safety at a local level. The project has focused on developing and strengthening local capabilities to produce homologated crime data on 11 types of incidents. Some efforts to use crime data to support policy-oriented crime research by CEASCI are taking place. A major weakness is the lack of trained professionals in the fields of sociology of crime and criminology according to the perspective of the ET. This is a major barrier to conduct research on the causes and consequences of violence, the role of place, and community-based crime prevention. A diploma in criminalistics and criminology was provided through UNAH for 120 officials of the Municipal Observatories for Citizen Coexistence and Security (OMSC) in 2016. GUATEMALA Guatemala is where InfoSegura has had the greatest impact in supporting the development of evidence-based public policies for citizen safety at a regional-departmental level. The project has focused on developing and strengthening local capabilities to produce homologated crime data on eight types of incidents. Political problems beyond the control of the project have impeded the implementation of some of the initiatives supported by InfoSegura aimed at improving the quality and availability of data on crimes, for example, the implementation of the Statistical Analysis Unit of MINGOB. In order to press forward toward achievement of the result of the Statistical Analysis Unit, the working strategy was altered, and the Third Vice Ministry hired two technical liaisons to work with InfoSegura. A COMMON CONCLUSION FOR THE 3 COUNTRIES It is important to acknowledge that the project strategy has responded to national priorities assigned by the institutions. Consequently, the starting point has been to address data requisition and creation of tools (i.e. PESS targeting and PESS monitoring, Chimaltenango Departmental Observatory Model). 16 4.3 EQ1 RECOMMENDATIONS EL SALVADOR There is a need to develop a crime and violence statistical system in El Salvador under the leadership of a specialized institution. DIGESTYC (or a future national statistics institute) is an obvious choice. But given its technical and financial weaknesses it is not a viable option. Also, it is not politically feasible for the MPJS to exercise leadership over such a system. A new statistical system could be placed under the leadership of a new specialized agency, similar to the Crime and Justice Institute of the U.S. whose responsibilities include all aspects of policy-oriented crime research. Such an institution could temporarily produce crime statistics until the system is fully developed and transferred to DIGESTYC. Several crime victimization surveys have been conducted by think tanks, newspapers, and private data collection agencies over the last 30 years in El Salvador. None can be considered an official source of crime victimization data. The DIGESTYC survey is the first official source of data on the topic and the unit record data should be made publicly available to allow its use by researchers and public policy analysts as should the data files of the Violence Against Women Survey. The lack of structured training on the production of crime statistics, its use for crime analysis to support police operations and crime investigation, and its further use to develop tools to support evidence-based citizen safety public policies is a major issue requiring InfoSegura’s attention. The need to provide high quality training on the use of crime data for different purposes requires the development of a specialized government structure in the fields of official crime statistics, crime analysis, and crime research. The proposed specialized institution mentioned above could assume leadership over the design of the specialized educational and training system. Existing institutions such as the National Academy for Public Safety and the Penitentiary Academy within the MPJS, the Attorney’s School within the FGR, any other specialized training body within the government structure, and public and private tertiary education institutions should be included as part of this initiative. UNDP might consider the inclusion of a criminology approach in the future, which might greatly enhance the work InfoSegura is undertaking on crime. A criminological perspective should not be seen as divorced from the UNDP’s paradigm for citizen security. This perspective provides the theoretical background for a view of citizen security as a democratic citizen order that eliminates the threat of violence in the population and allows for peaceful coexistence. Criminology is the scientific study of crime, including its causes, responses by law enforcement, and methods of prevention. As a subdivision of the larger field of sociology, criminology draws on psychology, economics, anthropology, psychiatry, biology, statistics, and other disciplines to explain the causes and prevention of criminal behavior. Criminology is more than a generalized, macro approach to citizen security. A criminological perspective provides explanations of the causes and consequences of violence at the macro, meso and micro levels. 25 25 Criminology offers explanations of criminal behavior and victimization that go beyond the macro approach. Some of these explanations include: 1) anomie-cultural learning theories to understand crime at the macro level; 2) social disorganization and control, and collective efficacy perspectives to community crime; 3) routine-activities and 17 A criminological perspective addresses the whole crime and violence problem in ways that go beyond reducing crime rates. Crime is a social phenomenon that affects the society as whole and touches the lives of a country’s residents. Crime prevention and control are two sides of a same coin. The development of information systems to support evidence-based public policy requires specialized criminological knowledge. Also, it requires the development of strong skills in fields such as statistics, geography, forensic sciences, and policing just to mention a few. An information system is just a key small piece of a comprehensive system aimed at improving the state capabilities to generate conditions to improve the population's quality of life, crime prevention through community action, accessible, agile and effective justice, education that is based on values, respect for the law, tolerance and healthy coexistence. The project should seek to facilitate access of public security and police officers to comprehensive training in criminology and crime analysis. HONDURAS There is a need to provide high quality training on the use of crime data for different purposes, which requires the development of a specialized occupational structure in the fields of official crime statistics, crime analysis, and crime research. The University Institute of Democracy, Peace, and Security (IUDPAS) and/or the National Autonomous University of Honduras (UNAH) could assume leadership over the design of the specialized educational and training system. Training could be provided locally through agreements between IDPAS-UNAH and municipalities. GUATEMALA In light of the ongoing political uncertainties in the country, the UNDP in its role as an inter￾institutional and impartial articulator should continue to seek to consolidate InfoSegura’s achievements thus far and strengthen the important institutional alliances surrounding data generation. The InfoSegura team should continue building up the counterpart’s technical opportunity-based approaches to understand crime, criminals and victims; 4) environmental criminology to understand crime at places and to develop situational crime prevention. Similar to the citizen security paradigm, criminological approaches put the emphasis on people in their roles of both actual and potential offenders and victims. Sound public policy rests upon knowledge obtained from comprehensive information systems on: what crimes occur, when and where do they occur, on whom, how do they happen, by whom are they committed, etc. Crime mapping, crime analysis, crime trajectory analysis, criminal careers, and the study of the responses to crime by both the state and citizens are a few examples of the many issues dealt with by the field of criminology. Understanding the fear of crime is another important policy issue. Fear and its associated insecurity, is the largest cost that crime imposes on citizens. Criminology offers explanations to the fear of crime, and criminological research abounds in findings on the most successful ways to deal with this problem. Prisoner treatment for rehabilitation and reintegration is another field for which criminological theory and research have advanced both at the policy and practice levels. 18 capacities for generating statistical knowledge products, which will provide some sustainability to the processes implemented. 4.4 EQ2 FINDINGS 4.4.1 EQ2: To what extent have Central American governments used (as planned) crime and violence data—produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions? EL SALVADOR InfoSegura has made a significant contribution to the formulation of citizen security￾related policymaking in the diverse forms of support it has provided in developing PESS and its monitoring system and in developing the index to prioritize municipalities to be included at each of the three phases of PESS. The First Monitoring Report on the PESS 2016-2017, published in August 2018, confirmed InfoSegura’s robust methodology and its utility for improving policymaking in the citizen security field. The significance of this contribution should not be underestimated. PESS, considered to be El Salvador’s most comprehensive and ambitious policy intervention on citizen security to date, was developed between 2015 by the CNSCC, an inter-institutional body constituted by government institutions, churches, the media, civil society, international donors, and multilateral organizations. A (US$)2 billion, five-year plan to combat crime and prevent violence, with interventions focused in the 50 most violent municipalities, PESS is considered to be a solid and comprehensive effort recognizing the structural factors underlying violence and containing a strong prevention dimension with comprehensive programs in education, health, and employment. It was constituted around five principal areas, 26 with 21 results and 133 short, medium, and long-term actions. Formulated with a results-based methodology, it was developed within a framework of actions, indicators, timeframes, goals, and required resources. The importance of the PESS monitoring system lies squarely in its capacity to contribute to the operationalization of PESS. Its efforts are focused on the collection, systematization, and analysis at the level of results and government management. Specifically, it measures the results and effects for short, medium, and long-term goals;27 the prevalence and incidence of crime for households and household members; the types of crime, victims, and victimization context; citizen perception of insecurity, victimization, satisfaction level with state services, and the strategic actions undertaken under PESS; 28 and processes of institutional strengthening for those participating entities. This methodology underpinned the First Monitoring Report on the PESS 2016-2017, published in August 2018. It provided key findings covering the five main strategic lines of action, including violence prevention, criminal control and prosecution, rehabilitation and social insertion, attention to and protection of victims, and institutional strengthening. The categories show differing degrees of progress due to a range of factors, pointing to a number of constraints the 26 The five principal areas were violence prevention, crime control and law enforcement, rehabilitation and reinsertion, witness attention and protection, and institutional strengthening. 27 CNSCC. 2017. Sistema de Monitoreo del Plan El Salvador Seguro. El Salvador. 28 This information is derived from the 2017 Survey on Victimization and Perception of Insecurity/Culture of Peace Survey. 19 PNC faces in order to consolidate results achieved thus far regarding short, medium, and long￾term goals. Key for policymaking, the report also provided concrete recommendations for overcoming these constraints. As a key input to PESS and its monitoring system, InfoSegura developed an Index of Municipal Prioritization,29 which provided key tools for characterizing the municipalities experiencing the highest levels of insecurity and vulnerability. This data system underpinned the decisions of the DIA and MJSP regarding the targeting of the most vulnerable groups in the 50 municipalities for the principal phases of PESS. The Ministerio de Justicia y Seguridad Pública (Ministry of Justice and Public Security, or PREPAZ) also used the results derived from this system to generate diagnostics and plans for the implementation of PESS in 24 municipalities for Phase III of PESS. Reflecting InfoSegura’s commitment to make gender one of the centerpieces of the project, it has generated an important and groundbreaking body of data and analysis regarding violence towards women in El Salvador that is informing policymaking in the area. According to InfoSegura’s project extension document, 30 the first phase of InfoSegura made explicit the existing limitations, weaknesses, and gaps in the processes required for the collection, generation, and analysis of citizen security and gender data. In order to better inform policymaking on gender, crime, and violence, InfoSegura supported a large number of studies on violence against women, the majority published in 2017, reflecting this increased focus on gender, particularly in the El Salvador program. These included the National Survey on Violence Against Women; Report on Incidents of Violence Against Women LEIV 2016-2017; Migration, Human Trafficking, and Access to Justice in the Context of Violence; and the National Information System on Human Trafficking in El Salvador. Also, in 2017, InfoSegura supported the underlying research for the document The Continuum of Violence Against Women in the Central American Region, which sought to provide key analysis and data in the search for policy consensus to improve current policies and models of comprehensive care for women in the countries in the program. Although all these documents offer important data for policymaking, the Report on Incidents of Violence Against Women LEIV 2016-2017 (based on data from the National Survey on Violence Against Women designed by InfoSegura and data from administrative records) concretely responds to legal obligations incurred by the Salvadoran State through the approval and entry into force of the Special Comprehensive Law for a Violence-Free Life for Women (LEIV). This law reflects the state’s recognition that violence against women is a public and structural problem that obliges the state to committed intervention given its grave impact on women, families, and society in general. The law implies an institutional commitment to obtaining the tools and instruments which will orient public institutions’ actions to identify, record, systematize, and make visible violence against women. According to LEIV’s mandate, measuring violence against women is the institutional responsibility of the MJSP. At present, the DIA operates the necessary system of indicators, which permits it to produce this required—now biannual—report. The second report reflects the participation of a wider number of agencies than the earlier report, allowing a more inter-institutional analysis for the years 2016-2017. It made use of administrative registries, which reflected data on a variety 29 Indice de Priorización de Municipios: herramientas para la priorización y focalización de políticas en materia de seguridad ciudadana. 30 P. 36. UNDP. nd. InfoSegura Project Extension Document. San Salvador, El Salvador: UNDP. 20 of aspects of violence, including violence in public and private spaces, sexual violence, femicide, violent deaths, and the majority of the incidents typified in the LEIV. InfoSegura’s activities in support of establishing a monitoring process of the SDG 16 agenda will permit SETEPLAN to generate key inputs to security sector policymaking focused on achieving this goal. In 2015, El Salvador was selected as one of 15 countries in the world to implement a pilot initiative to strengthen the monitoring framework for the United Nations (UN) SDG 16 (Peace, Justice, and Strong Institutions). Supporting SETEPLAN, InfoSegura participated in a process with civil society, the private sector, and a range of public institutions to validate a series of indicators and data sources for the monitoring of the goal, which would oversee security policies through 2030. This data will be available through the DIA’s web portal. InfoSegura supported the development of a methodology for an Índice de Priorización de Centros Educativo en Materia de Seguridad Ciudadana (Prioritization Index on Educational Centers Related to Citizen Security Issues, or IPCESC). The IPCESC methodology for the prioritization of educational centers serves to identify those schools that are most affected by citizen security issues in order to formulate appropriate policies. This information provides the key institutions with timely and reliable information to appropriately target those schools most affected or at risk in order to develop plans of action to minimize or eliminate threats. Using this methodology, MINED and the PNC, in coordination with the Vice Ministry for the Social Prevention of Violence and the Municipal Committees for the Prevention of Violence, identified 250 schools located in 26 of the prioritized municipalities covered by PESS. An early warning system was devised in order to provide the police with the ability to react in a timely manner to crisis situations. The process of identifying the schools underscored that, in addition to the results of the indexing, the knowledge of local actors is key in designing effective interventions. The school index has not been made public to avoid the stigmatization of specific schools and communities. HONDURAS InfoSegura has made a significant contribution to the formulation of citizen security￾related policymaking in its development of a Municipal Prioritization Index. The Honduras government made the decision to have a decentralized system for the production of crime statistics and the development of crime prevention strategies. National data is produced by aggregating local data from the OMSCs. The Municipal Prioritization Index developed with InfoSegura’s support was instrumental in identifying the municipalities where the OMSCs were established. Crime prevention initiatives are being developed and monitored at the local level. This is a significant achievement as it allows for the effective involvement and consequent empowerment of local actors in the identification of strategies for the dissuasion and control of crime, and the implementation of promising social and situational crime prevention initiatives. The report, entitled Del Dato a la Acción: Todas las Historias Cuentan (From Data to Action: All the Stories Count) was jointly produced by the Secretariat for Safety, UTECI, and the FGR with project support of InfoSegura. The report documented actions and municipal ordinances that have been effective in diminishing the levels of violence at the local level in 12 municipalities. 21 GUATEMALA InfoSegura has made a notable contribution to the formulation of citizen security￾related policymaking in Guatemala through its contributions to the National Strategy for the Prevention of Crime and Violence 2017-2027 (ENP), including developing the index to prioritize the municipalities to be targeted for interventions and identifying the key locations for departmental crime observatories (pilots in Chimaltenango and Guatemala City). InfoSegura supported both the implementation and validation of the Municipal Index. This index permits the government’s operationalization of both the ENP and its Citizen Security Program. InfoSegura also provided support for the installation of two automated tools for decision-making, including an advanced analyzer and an automated report tool. It played a key role in developing the model of observatories, which in turn will be an important tool in designing policies tailored to local realities. InfoSegura provided key support for the National Survey on Perceptions of Public Security and Victimization (ENPEVI), developed with both MINGOB and INE, which will provide crucial data to underpin policy decisions. This survey received support jointly by the Center for Excellence, MINGOB, INEGI, UNODC, UNDP, as per inter-agency agreement between UNDP and UNODC. The survey is a key component of the ENP and the first of its kind, covered 22 departments over the period from November 2016 to October 2017. ENPEVI adopted the recommendations of the Latin America and the Caribbean Crime Victimization Survey Initiative as well as the International Classification of Crime for Statistical Purposes. Its results permit reporting on the progress of four indicators of the 2030 SDGs, particularly Goal 16. The results were presented at the governance roundtable constituted by MINGOB, INE, INEGI, and UNODC. InfoSegura developed a support document to strengthen results analysis of the ENPEVI and the capacities within the Vice Ministry. In October 2018, personnel of the Vice Ministry of Violence and Crime Prevention participated in a workshop on data exploitation of the ENPEVI at the Center of Excellence for Statistical Information on Government, Crime, Victimization, and Justice (COE). The workshop sought to provide the GOG with technical support to generate secondary indicators for the analysis of ENPEVI data. The principal challenge at this point is the publication of the ENPEVI, the delay of which is attributable to the tense and volatile political situation in Guatemala, intensified by the upcoming electoral period in 2019. InfoSegura contributed to the development of a Model for Departmental Observatories on Citizen Security and Violence Prevention in Guatemala, permitting the launching of pilot observatories in Chimaltenango and Guatemala City. This institutional model, which constitutes a major contribution to policymaking in the citizen security sector, will provide objective, reliable, up-to-date data to inform departmental plans and strategies to more effectively confront crime and violence and their causal factors. The observatories are broadly under the purview of MINGOB, and under the technical supervision of the Statistics and Strategic Analysis Unit. They will eventually operate in every department in the country and in their advisory councils on security and prevention. 22 The general objectives of the departmental observatories include:31 1) obtain updated diagnostics on violence and criminality in each of the departments in the country (promotion of standardized reports); 2) stimulate and strengthen the practice of articulation, exchange, analysis, and dissemination among the relevant actors regarding public policies on security and prevention in each of the departments (promotion of the methodology of the inter-institutional technical roundtables; and 3) institutionalize practices of M&E of preventive polices based on evidence. 4.4.1.1 EQ2.1: What are the internal and external factors that foster or impede the governments to use the data? EL SALVADOR Factors that serve to foster the use of data are existing policy frameworks and legislation for which data is key. The PESS framework has made it possible to develop and tailor data to a range of specific policy needs and requirements. The process behind the drafting of PESS, underpinned by extensive inter-institutional and inter-sectoral discussion and debate, facilitated a more fluid use and exchange of data than in the other two countries where the institutional interfacing was more challenging. InfoSegura has sought to strengthen these processes of institutional cooperation through promoting bilateral institutional agreements, such the one between the MJSP and the Ministry of Health. A factor that impedes the greater use of the data for policymaking is the insufficient analytical capacity in many STUs and source institutions for this data. This is not merely an observation by the ET but is a clearly voiced concern by public officials at a variety of levels in distinct institutions, including within the PNC. While there is an appreciation and understanding of the importance of the data being generated for evidence-based policymaking, there are simultaneously capacity gaps in terms of taking full advantage of the wealth of data being generated. Further, as there is an increased focus on multidimensional analysis on citizen security and development, the challenge of translating this material into effective, concrete policies is greater. A challenge for InfoSegura in its upcoming period might be to seek to devise strategies to strengthen institutional analytical capacity. In the case of the LEIV, the law actually mandates the production and use of the data, which clearly improves its prospects for use. Perennial issues that threaten the ongoing use of the data for public policymaking are staff rotation/turnover and changes in public policies. Staff and policy changes after elections are common and, in El Salvador’s case, there are a number of unknowns looming on the horizon. Most informed observers, however, believe that PESS, or most of its most important features, will probably be largely retained as a policy, and that the well-embedded data systems will not be disrupted. An additional impediment in the use of data for evidence-based policymaking is the weakness of some of the most relevant public institutions linked to these efforts – in the case of El Salvador, the weakness of DIGESTYC. In some areas, DIGESTYC’s technical shortcomings adversely affect the institutionalization of harmonized data produced under InfoSegura. Ideally, DIGESTYC would play a more robust role in InfoSegura, contributing to the sustainability of results, but this key institution’s limitations do not permit it to fully fulfill its 31 InfoSegura. 2017. Modelo “Observatorios departamentales de seguridad ciudadana y prevención de la violencia en Guatemala”. Guatemala. 23 mandated role in as timely and effective a manner as required. There is a lack of clear and institutionalized mechanisms or guidelines for the channeling of information to policymakers. The ET detected a number of cases in which policymakers were not fully aware of the extent and breadth of the data available. HONDURAS An intensely polarized political situation gave way to a complex election context in 2017 which brought major policy changes in the area of citizen security, shifting the policy orientation towards the local level. A factor which favored the use of InfoSegura data is that the project participated in months of roundtables and discussions with the Sub-Secretary for Security to define the specific necessities of data in methodological terms and within the new policy framework for crime and violence prevention. Participating in these key stages favored the use of the data generated as it was calibrated carefully to the developing broader policy structure. Some elements that impede the use of data for policymaking are linked to the lack of trained professionals in fields related with the sociology of crime and criminology. According to an informant, this is a major barrier to conduct research on the causes and consequences of violence, the role of place, and community-based crime prevention. A positive note, however, is CEASCI’s current effort to use crime data to support policy-oriented crime research in a multidimensional study. Insufficient capacity for analysis in relevant government entities impedes the use of data for policymaking. Capacity deficits in data analysis within government agencies that use data on crime and crime prevention constrain their ability to benefit from this data. GUATEMALA The highly volatile and fragile political situation in Guatemala has created a disrupted institutional environment in general and the nature of the crisis can reverberate throughout state institutions. However, InfoSegura continued to work together with other institutions to improve the cycle of generating statistical data as a whole, at the national level from the Interinstitutional Technical Roundtable. The Ministry of Governance is discussing results analysis within the ENPEVI governance roundtable, composed of INE and MINGOB. Nonetheless, political sensitivities have delayed the publication of the ENPEVI to date. The insufficient capacity for data analysis in different state entities was cited by government staff as an impediment to using data for policymaking. The participation of personnel of the Vice Ministry of Violence and Crime Prevention in a workshop in Mexico on ordata exploitation of the ENPEVI at the COE is an example of a useful step to improve analytical capacity. 4.5 EQ2 CONCLUSIONS EL SALVADOR InfoSegura’s support has provided key inputs for policymaking in the citizen security sector through: 1) its contribution to PESS; 2) its provision of municipal prioritization indices that permit targeted interventions and policies tailored to the needs of these areas; 3) its robust design of a PESS monitoring system, which allows policy M&E; and 4) its monitoring report, which provides reliable and concrete data on progress and ongoing challenges. InfoSegura has generated an 24 important and groundbreaking body of data and analysis regarding violence towards women in El Salvador that is informing policymaking in the area and already fills a legislative requirement. The project’s support to developing a monitoring system for Goal #16 of the SDGs will contribute to shaping policymaking in order to achieve this objective. A number of factors foster or impede the process. These include the existing policy frameworks and legislation for which the data is key, as in the case with the LEIV law discussed above. The PESS framework, in particular, has made it possible to develop and tailor data to a range of specific policy needs and requirements. A factor that impedes the greater use of the data for policymaking, and which is also evident in the other two countries, is the lack of analytical capacity in many STUs and source institutions for this data. HONDURAS InfoSegura has contributed to an effective use of data and information to inform evidence-based policymaking at the municipal level. InfoSegura’s support to the Municipal Prioritization Index has enabled the identification of 30 municipalities for the establishment of OMSCs. Under the leadership of UTECI, and with InfoSegura’s support, local governments are using crime data to develop and monitor crime prevention initiatives. The insufficient capacity for analysis in different state entities which could be making use of available data impedes the more effective use of data for policymaking in Honduras. GUATEMALA InfoSegura has contributed to a qualitative leap in data and information use to inform evidence￾based policymaking. InfoSegura’s support to the Municipal Prioritization Index, which underpins strategic interventions in the ENP, has also been used to establish pilot observatories like the one in Chimaltenango Department and has been a key underpinning for the ENP. The uncertain political situation in Guatemala, which can affect multiple state institutions, poses a potential impediment to some of the processes put in place with InfoSegura’s assistance and the government’s use of data for policymaking in citizen security and violence prevention. Although political sensitivities have delayed the publication of the ENPEVI, analysis of results continues to be discussed within the framework of the governance round table, and discussion of the issue of making the data official is underway at INE, the body responsible for publishing official data per survey. However, ENPEVI data collection was carried out in February of 2018 and its publication has now been delayed for over a year. Observers have noted that if ENPEVI is not published soon, the effort runs the risk that the data and infographics produced will lose relevance. Government staff in different institutions, including the PNC, noted that the insufficient capacity for analysis in different state entities which could be making use of available data impedes the more effective use of data for policymaking in El Salvador. 4.6 EQ2 RECOMMENDATIONS EL SALVADOR Considerable use has been made of data in policymaking as evidenced by InfoSegura’s role in the monitoring of PESS among other areas previously discussed. However, there needs to be greater focus on the enhanced use of the data found in the DIA web portal, which represents a notable 25 advance in access to updated information on the critically important, yet historically neglected, issues of violence against women and human trafficking. HONDURAS Since Honduras has made such important strides in the use of data for policymaking at the local level, InfoSegura should identify the areas in this institutional architecture that can further strengthen and consolidate the work of UTECI and the other STUs. InfoSegura should further support CEASCI’s efforts to use the multidimensional study it is completing as an input to evidence-based policymaking. GUATEMALA As with the other countries of the Northern Triangle, there is a need to strengthen analytical capacity within relevant public institutions for policymaking in law enforcement and violence strategies. InfoSegura should seek to strengthen and optimize the institutional architecture for information management for the Departmental Observatories, providing key technical and other support to the new pilot observatory in Chimaltenango. Continued support should be provided to the process of the “Certification of Statistical Quality” for the PNC in Guatemala, being undertaken jointly by the Center for Excellence, MINGOB, INEGI, UNODC, UNDP, as per inter￾agency agreement between UNDP and UNODC. 4.7 EQ3 FINDINGS 4.7.1 EQ3: To what extent has the creation of partnerships and transparency (CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy? EL SALVADOR, GUATEMALA, AND HONDURAS InfoSegura has made important strides in overcoming the historical lack of trust between government and civil society, which has, in specific instances, led to collaborative relationships between CONOSE institutions (and civil society) and governments. InfoSegura’s program documents indicate that think tanks, academia, and civil society were envisioned from the project’s outset to play a role in all four phases of InfoSegura’s generation of information for public policy formulation, including production, analysis, use, and dissemination. A Regional CONOSE32 was established as a key mechanism to achieve this goal, constituted by academia, research institutes, and civil society involved in citizen security issues, with InfoSegura serving as a bridge and regional platform. Regarding one of the contextual challenges identified in the original program document, that of the historical distrust between governments and civil society, there is evidence that progress has been made, principally in both El Salvador and Honduras. FUNDAUNGO, a think tank and 32 Members of CONOSE: Guatemala: Facultad Latinoamericana de Ciencias Sociales, (FLACSO); Sede Académica Guatemala; Universidad Rafael Landívar (URL). Honduras: Asociación para una Sociedad más Justa (ASJ); Proyecto FLACSO Honduras, UNAH. El Salvador: FUNDAUNGO; Universidad Centroamericana “José Simeón Cañas” (UCA); FLACSO Programa El Salvador. Costa Rica: FLACSO Sede Académica; Universidad de la Paz (UPAZ). Panama: FLACSO Programa Panama. Dominican Republic: FLACSO Programa República Dominicana. 26 member of CONOSE, provided key technical assistance for developing the methodology33 for the monitoring of PESS. Given the dimensions and importance of PESS, 34 FUNDAUNGO’s technical responsibility in developing the monitoring system is an indication of the level of professional confidence it enjoys among public policymakers. 35 This rigorous, results-based methodology generated the First Monitoring Report on the PESS CNSCC2016-2017, which provided detailed, territorially-focused information and was presented to the Salvadoran President in August 2018. A further example of civil society/government collaboration in El Salvador relates to the UN 2030 Agenda. InfoSegura developed a monitoring system on progress on the implementation of the SDGs for Goals 5 and 16, which was then adopted by SETEPLAN and validated in sessions with civil society and the private sector. In Honduras, a CONOSE member, Association for a More Just Society (ASJ) developed a series of baseline studies in 2016 that have been utilized by the Ministries of Security, the Public Ministry, and the Judicial Branch.36 Conversely, in Guatemala, relations between government and civil society have been severely strained over recent years and collaboration has been limited. InfoSegura has undertaken a range of diverse initiatives in the areas of regional fora, training activities, conferences, and workshops, reflecting a recognition of the need to strengthen training and research capacities in the region that would reflect both civil society and governmental needs and perspectives. The CONOSE regional network pursued four principal strategic lines of action to support analysis of the citizen security challenge: 1) develop the capacity of academic institutions in the region and the establishment of collaboration mechanisms, raising the quality standards in knowledge and information generation; 2) strengthen the training opportunities to improve human capital to understand and respond to the problem; 3) promote the design and implementation of a regional research agenda; and 4) contribute to evidence-based, informed debate, including spaces for dialogue and reflection with representatives of governments. Toward this end, InfoSegura has set in motion a process, which has included regional fora, national seminars, and dialogue events, among a range of other activities. To launch this process, in 2016 InfoSegura sponsored the first CONOSE Regional Meeting on Citizen Security aimed at building an alliance between the different think tanks, foundations, and universities focused on strengthening training and evidence-based research opportunities in citizen security. Follow-up to this initiative resulted in the 2017 First Regional Forum of the Regional Knowledge Network 33 The development of the monitoring system involved a number of different actors such as the Governance Secretariat and the Vice Ministry of Social Prevention of the Ministry of Justice and Public Security, UNDP, and the Technical Secretariat of the CNSCC. The German Agency for International Cooperation (GIZ) and the Dirección de Innovación Tecnológica e Informática del Gobierno de El Salvador (Directorate of Technological and Computer Innovation of the Government of El Salvador, or ITIGES) also participated in the development of the information platform. 34 PESS was developed in 2015 by CNSCC, an inter-institutional body constituted by government institutions, churches, the media, civil society, international donors, multilateral organizations, and civil society. A (US$)2 billion five-year comprehensive plan to combat and prevent violence, its principal prevention measures are interventions in the 50 most violent municipalities. 35 It should be noted, however, that the role of civil society and academia in contributing to public policymaking in El Salvador preceded InfoSegura in that there have been roundtables for dialogue on these issues for a number of years. 36 The ASJ developed a Performance Measurement Index (Baseline Index) for the three institutions noted above. 27 on Citizen Security in Tegucigalpa at UNAH, attended by the 12 members of CONOSE and government representatives from the region, UNDP, and USAID. In an effort to improve the quality of research on crime and violence prevention in the region, one of CONOSE’s stated goals, the Second Regional CONOSE Forum reflected the more focused interest by the CONOSE network to address the challenges of conducting fieldwork in violent contexts. Lead efforts in this area of activity were undertaken by CONOSE colleagues from the State University of Rio de Janeiro Laboratory of Violence Analysis (LAV-UERJ) and the Central American José Simeón Cañas University (UCA) through its University Institute for Public Opinion Research (IUDOP). These efforts led to the production of the Technical Protocol and Field Research Toolkit for Working in Violent Contexts. The technical protocol was disseminated in a series of workshops in each of the three Northern Triangle countries as part of InfoSegura and CONOSE’s strategy to strengthen research capacities of civil society, academics, and government representatives. A further initiative to strengthen training opportunities in the region has been the development of the regional and institutional virtual Central America Diploma in Citizen Security and Violence Prevention, developed by the Facultad Latinoamericana de Ciencias Sociales (Latin American School of Social Sciences, or FLACSO) Costa Rica and directed at government staff working in the citizen security sector. The course, with 30 participants, was primarily focused on policy and project design, implementation, and evaluation of interventions in the area of citizen security and violence prevention. In June 2018, CONOSE mounted its Third Regional Forum: Violence, Coexistence, and Development: Advances and Challenges for Research and Policy Formulation on Security, which sought to leverage the incremental progress made with previous initiatives to further enhance the discussion and debate among academia, civil society, and government representatives regarding the incidence of violence in the different development spheres. The forum’s presentations covered progress in understanding the phenomenon relevant to policymaking, as well as ongoing tasks and challenges that persist in the realm of research. 4.7.1.1 EQ3.1: What are the internal and external factors that foster or impede the collaboration among the members of CONOSE? EL SALVADOR, GUATEMALA, AND HONDURAS A variety of impediments continue to prevent CONOSE from fully achieving its principal goals. Missing to date is a joint research agenda between state and civil society, which would eventually result in mutually enriching, concrete, and effective collaboration in problem￾solving between the two sectors. Another key challenge is that the necessary dynamism fueling successful network structures, especially ones that seek to combine state and non-state actors sharing common interests, has yet to emerge for CONOSE. Government officials and the academic community, while appreciating and participating in CONOSE events, appear to operate along parallel lines without a shared framework, vision, or definitions of violence and crime prevention. There are still significant gaps that exist between government needs for evidence￾based studies on the efficacy/effectiveness of specific crime prevention initiatives/strategies versus the tendency in academia to pursue more theoretical concerns. The absence of research within the framework of criminology and the absence of training opportunities in criminological approaches is yet another key impediment underlying this gap 28 between governmental needs for evidence-based studies and research. This breach between research needs and dominant research tendencies serves to undermine the kind of commitment to and sense of common interests and goals that would cement a more robust collaboration between public policymakers in the citizen security/law enforcement sector, academia, and civil society analysts. A final impediment to CONOSE is that its network structure and lack of legal status compromise access to broader funding opportunities, which are critical for chronically underfunded research institutes and universities. Notwithstanding, discussions with members and observers of the CONOSE network suggest that it has made significant strides in opening new spaces for inter-institutional dialogue and in promoting awareness of the ongoing need for shared approaches to crime and violence prevention. The various forums that have been held provide a unique opportunity for interaction between state and non-state actors in the citizen security field, actors who rarely have other venues to meet and participate in exchanges on issues of mutual professional interest. 4.8 EQ3 CONCLUSIONS The CONOSE network has made important strides in overcoming the historical distrust between governments and civil society regarding possibilities for collaborative efforts in joint problem￾solving and has undertaken efforts to both: 1) build stronger inter-institutional relationships at a regional level; and 2) strengthen civil society and academia’s capacity for research and training. Nonetheless, important challenges still exist with respect to the construction of a joint research agenda between academia and government policymakers in the citizen security sector. 4.9 EQ3 RECOMMENDATIONS If CONOSE is to fulfill the mandate originally envisioned for it, and generate effective inter￾institutional collaboration, it will need to take steps to ensure that the research activities increasingly respond to a common agenda between government officials and academia. It would be useful to hold a forum specifically focused on establishing this agenda and exploring the current gaps that exist between government needs for evidence-based studies on efficacy/effectiveness of crime prevention initiatives/strategies versus the tendency in academia to pursue more theoretical concerns. In order to respond to the existing lack of academic opportunities in the region for training in criminology, it would be useful to examine the possibilities of collaborating with universities and experts in the hemisphere to expand curricula to include such courses as well as inviting select experts to conduct seminars and courses in this area. Although the evaluation team notes that the efforts presently underway between John Jay School of Criminal Justice and the UCA to develop a master's degree in criminology and accreditation could contribute to tackling this gap, this is a long-term issue. Given the need for funding research initiatives, InfoSegura needs to explore mechanisms by which CONOSE could receive outside funding given the obstacle created by CONOSE’s loose network structure and lack of legal status as an institution. 4.10 EQ3 GENERAL RECOMMENDATIONS Training programs in the fields of official crime statistics, survey sampling, crime measurement scales, crime analysis for crime investigation and police operations, and evaluation of crime prevention policies and programs should be designed and implemented in the short term. 29 InfoSegura should strengthen analytical capacity for evidence-based policymaking within relevant institutions to ensure the data available is fully exploited. With an increased focus on multidimensional analysis of citizen security and development, the challenge of translating this material into effective, concrete policies is even greater. Given the threats to the sustainability of results due to major electoral changes or ongoing political volatility, the project should seek to strengthen institutional relationships through formal agreements, maximizing institutional commitments to policy frameworks and information sharing to consolidate the institutional architecture and spaces established by the project. In its upcoming final phase, InfoSegura should identify priority areas for consolidation of critical areas of work in the three countries, rather than launching diverse new activities, which may result in dispersion. Identifying the principal weaknesses that serve as obstacles for the effective realization of core goals could provide a road map for concentrating efforts. As noted in the report, and particularly strongly evidenced in the case of El Salvador, InfoSegura generated significant data and analysis on gender and violence against women. Thus, it is recommended that “if it is still possible, a knowledge management systematization of the activity should be conducted to identify lessons learned and good practices applied regarding the mainstreaming of gender in information management for citizen security.”37 37 P. 26. 2018. Monitoring and Evaluation System Assessment of USAID Regional Security Activity InfoSegura. San Salvador, El Salvador. 30 ANNEXES 31 ANNEX I: EVALUATION SCOPE OF WORK STATEMENT OF WORK (SOW) USAID/CAM Final Performance Evaluation of the Regional Citizen Security Activity (InfoSegura) A. Purpose of the Evaluation The purpose of the Final Performance Evaluation of the USAID Regional Citizen Security Activity (known as InfoSegura) in the northern Central American Triangle region (El Salvador, Honduras, and Guatemala) is to document activity achievements and challenges to date, inform USAID of any adjustments for the remainder of program implementation, and help with appropriate planning for future citizen security work. In addition, the evaluation will provide empirical evidence on management issues and support learning and continuous improvement in USAID’s regional citizen security work for current and future activities. The principal audience of this final performance evaluation within USAID/CAM, which is the Mission managing the regional activities, will be Mission Management, the Democracy and Governance Office, and the Regional Program Office. Other USAID offices with interest in the evaluation results are the Democracy and Governance Offices within USAID/Guatemala and USAID/Honduras; the Bureau for Latin America and the Caribbean (LAC) (particularly the team who coordinates the Central America Regional Security Initiative); and the Bureau for Democracy, Conflict and Humanitarian Assistance. The implementing partner, the United Nations Development Programme (UNDP), will carry out the remaining activity implementation based on findings and recommendations from the evaluation. USAID/CAM will use the findings, conclusions, and recommendations of this evaluation to make any necessary adjustments for the remainder of InfoSegura implementation, as well as determine areas and approaches for future regional citizen security activities based on USAID/CAM strategic goal and development objectives. In addition, evaluation results will be used for reporting purposes to stakeholders: Central and local governments in the Northern Triangle, other U.S. Government agencies, other bilateral and multilateral donors working on citizen security issues, civil society organizations, and academia. The main participants in the evaluation will be the prime implementing partner (UNDP), activity beneficiaries, public institutions, academia, civil society organizations, and USAID. This Final Performance Evaluation is scheduled for the end of calendar year 2018, with approximately one and a half years left of implementation. The evaluation will cover the period from May 30, 2014 through September 30, 2018 B. Background Information about the InfoSegura activity As part of the Citizen Security and Good Governance component of the USAID Central America Regional Development Cooperation Strategy (RDCS), USAID/CAM signed a Public International Organization (PIO) agreement with UNDP on May 30, 2014. The activity is slated to end in May 2020. USAID/CAM has modified the agreement five times to adjust activities and results, and to increase the total funding amount from approximately (US$)12,000,000 to (US$)21,000,000. 32 Activity Implementing Partner Award Number and Dates Funding Regional Citizen Security Activity - InfoSegura UNDP AID-596-IO-14-00001 May 30, 2014 – May 31, 2020 (US$)21,000,000.0 0 Within USAID’s RDCS 2015-2019, InfoSegura contributes to Development Objective 3 “Regional Human Rights and Citizen Security Improved.” Specifically, it contributes to intermediate result (IR) 3.1 “Regional capacity to address citizen security through more coordinated governance improved.” Under the U.S. Strategy for Central America (CEN Strategy), InfoSegura contributes to the achievement of the Goal Two “Enhance Security and Reduce Levels of Crime and Violence” of the strategy’s Pillar Two “Enhanced Security.” Within the Alliance for Prosperity in the Northern Triangle (A4P), InfoSegura contributes to the A4P Line of Action Three “Improving Public Safety and Enhancing Access to the Legal System.” InfoSegura was designed to focus on strengthening evidence-based public policy in citizen security in six countries: Guatemala, El Salvador, Honduras, Belize, Costa Rica, and the Dominican Republic. However, implementation efforts have focused on the three Northern Triangle Countries. InfoSegura responds to the persistent lack of analysis and understanding of crime and violence—which has resulted in historically poor crime data, analysis, and policymaking by governments. The lack of collaboration both by government and civil society to constructively engage to address these challenges has been another critical need. To address these challenges, the activity has worked along the chain of information management (information collection, analysis, use, and dissemination) to strengthen evidence-based policymaking on citizen security. This is achieved through: 1) improving regional and national capacity for collecting, monitoring, and processing information regarding security, and its use in decision-making and policymaking at both levels; 2) strengthening civil society capacity for collecting, analyzing, and processing information on citizen safety and monitoring policies and programs on citizen safety; and 3) supporting regional-level knowledge management to enhance the understanding of fundamental causes of violence and insecurity and to promote successful policies and practices. The design of InfoSegura included broadening partnerships through support for closer coordination between the General Secretariat of the Central American Integration System (SICA, by its acronym in Spanish), although this was dropped early in the program. National governments, civil society, and academic institutions, along with other key USAID bilateral and regional citizen security, migration, and human rights programs, have been key partners. The purpose of these partnerships was to build capacity of both government and civil society as major beneficiaries, while improving collaboration at the regional level. This helps to ensure that InfoSegura integrates with other activities (such as local-level crime prevention initiatives), or other critical regional themes (such as migration or human rights) to accelerate effectiveness. InfoSegura includes two components: 1. Evidence based policymaking on citizen security increased. 2. Regional collaboration, knowledge management, and networking on citizen security increased. 33 The two objectives under Component One are: Quality and comparable citizen security indicators consolidated. This objective will involve final investments at the country level to consolidate collection, monitoring, and systematization processes of crime and security data. Key activities include developing key citizen security indicators that are comparable and disaggregated and to develop and support sustainable institutional capacity building plans—particularly within the Sub-Technical Units of Security Ministries, charged with citizen security policy to become permanent operating structures. 1. Analytical capacity to Inform Citizen Security Policy Decisions Strengthened. This objective will seek to strengthen both analytical and institutional capacities of key public institutions and civil society networks to provide evidence-based analysis and policy recommendations to decision-makers on citizen security policy. The two objectives under Component Two are: 1. Improved dialogue between civil society and government institutions on citizen security issues. Under this objective, civil society networks and platforms (which include an InfoSegura-backed Red de Conocimiento sobre Seguridad Ciudadana – CONOSE) 38 are developed and supported at both the national and regional levels. These networks and platforms share analysis and advocate on major themes of citizen security, along with Alliance for Prosperity progress, with government partners, and within civil society. 2. Exchange of citizen security best practices and effective strategies increased. This includes the promotion of expert exchanges of, evidence-based tools, forums, and approaches from other countries to help Central America governments and civil society better tackle citizen security issues. Since its inception, the most significant technical modification to the InfoSegura activity has been the ending in 2015 of planned support to the Regional Security Commission of SICA to construct a regional crime observatory—called OBSICA. Because of the lack of consensus in the Regional Security Commission, InfoSegura could not support the development of the OBSICA Regional Observatory and could not implement activities in Nicaragua. USAID/CAM and the UNDP subsequently re-directed the activity to focus support to the remaining seven countries of SICA, emphasizing national level development and close regional collaboration between these countries—with specific emphasis on the Northern Triangle—rather than engaging any further with SICA. C. Evaluation Questions USAID/CAM has identified the following three main evaluation questions, in priority order. These should be answered by the Evaluation Team and clearly presented in the Final Report in terms of how they relate to the evaluation purpose. The Evaluation Team should answer the sub- 38 CONOSE is integrated by higher education institutions, think tanks, and non-governmental organizations of Central America and the Dominican Republic aimed to promote and improve institutional capacities, professional competencies, and knowledge generation based on evidence of citizen security and peace. More information in Spanish can be found in www.conose.org. 34 questions separately in the report and use the data collection and analysis of the sub-questions to answer the main questions. The Evaluation Team should incorporate, to the extent feasible, analysis of possible differences associated with gender or social groups, particularly historically excluded groups (youth, people with disabilities, indigenous populations, etc.), and they should be reported separately for men and women. As an example, the Evaluation Team should consider gender and social group￾disaggregated data when analyzing the capacity of the institutions to collect and use data. 1. To what extent has InfoSegura contributed to change the level of quality39 and timeliness of crime and violence data definition, collection, and recording in the region? 1.1 To what extent has InfoSegura developed the capacity within the Sub-Technical Units (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? 1.2 To what extent has InfoSegura developed the capacity of source institutions (such as the National Civilian Police, the Attorney General’s Office, and the Forensic Medicine Institute in El Salvador) to ensure quality of data and improve information sharing between source institutions and Sub-Technical Units? 1.3 What is the potential that the capacity developed within the Sub-Technical Units and the source institutions increases or decreases after the completion of InfoSegura? 2. To what extent have Central American governments used (as planned) crime and violence data— produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions? 2.1 What are the internal and external factors that foster or impede the Governments to use the data? 3. To what extent has the creation of partnerships and transparency (CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy? 3.1 What are the internal and external factors that foster or impede the collaboration among the members of CONOSE? D. Evaluation Methodology, Data Collection, and Analysis For the InfoSegura performance evaluation, a non-experimental mixed-methods design that combines a comprehensive, rigorous analysis of existing quantitative data with customized qualitative techniques designed to elicit primary data from a wide range of counterparts, partners, beneficiaries, and stakeholders is recommended. This approach allows for triangulation of complementary data to elucidate linkages between activity inputs, outputs, and outcomes. The Evaluation Team should consider a range of possible methods and approaches for collecting and analyzing the data to address the evaluation questions thoroughly. 39 Based on the best practices of the industry, the Evaluation Team needs to specify in the Evaluation Plan what criteria/methodologies/tools they will use to measure quality of crime and violence data and the justification for the selection of these criteria. 35 The use of participatory methods and activities that will enhance collaboration and dialogue among counterparts is required. Further, data collection and analysis methods should be sensitive to possible differences related to sex and/or social status and should follow applicable Institutional Review Board (IRB) guidance on data security to ensure safety and confidentiality of all individuals providing data or information for the purposes of the evaluation. The finalized evaluation method(s) and approaches, data collection plan, and analysis will be included in the Evaluation Plan submitted to USAID for revision and approval before field visits and data collection begin (see Deliverables section below). The method(s) proposed should comply with the USAID Evaluation Policy. The data collection plan for this evaluation will include, at a minimum: a desk review of relevant documents such as the cooperative agreement and work plans; review of activity performance monitoring and context data; key informant interviews and/or focus group discussions promoting equal participation of women and men; and direct observation through site visits. USAID/CAM expects both qualitative and quantitative data to be collected; and the results will be coded, triangulated, and analyzed for content. The Evaluation Team is encouraged to propose additional/alternate data collection and analysis methods in the Evaluation Plan that could yield stimulating, robust evidence in answering each of the evaluation questions. Data collection shall be systematic and data must comply with the five data quality standards of validity, integrity, precision, reliability, and timeliness. Specific interview, survey, and/or focus group protocols will be appended to the Evaluation Plan and finalized with approval from USAID/CAM; the questions should be used to answer the evaluation questions listed in this document and address the purpose of this evaluation. All data collected in response to the evaluation questions must have as much level of disaggregation as possible. As minimum, and per USAID Gender Equality and Female Empowerment Policy, all people-related data must be disaggregated and analyzed by sex, as well as analyzed for any differences between effects on men and women or male and female participation. Desk review of relevant documents USAID/CAM will provide the Evaluation Team with all relevant strategy and activity specific documents, such as the cooperative agreement between USAID and UNDP and its amendment(s) including the expected results, performance reports stating the results achieved, any prior assessments if applicable, etc. The Evaluation Team must review these documents and other existing literature provided by USAID and others in preparation for the initial team planning meetings and before meeting with local stakeholders for interviews. The Evaluation Team is expected to conduct a contextual literature research and review and create a Review Matrix to be delivered to USAID as part of the final Evaluation Report, indicating how key information extracted from reviewed documents and other methodologies were linked to each evaluation question. At minimum, the Evaluation Team shall review the following documents relevant to the InfoSegura Final Performance Evaluation: • Regional Development Cooperation Strategy 2015-2019 (Link here) 36 • InfoSegura Description and Modifications • InfoSegura Monitoring and Evaluation Plan • InfoSegura annual work plans • InfoSegura biannual reports • Evidence-Based Tools, Platforms, Studies, and Reports used to address policy challenges: Regionally and Nationally • CEN Strategy (Link here) • Plan of Alliance for Prosperity in the Northern Triangle (Link here) Review of performance and context indicators InfoSegura has an activity-specific Monitoring, Evaluation, and Learning Plan and has collected data on a number of standard and custom performance indicators during activity implementation. This monitoring data will provide one source of data on progress toward objectives and outcomes. The Evaluation Team will use monitoring data on performance indicators as part of the evaluation analysis and should report on it in the Final Report as much as it relates to the evaluation questions stated above and satisfies relevant data quality standards. Context data should be analyzed and included to the maximum extent possible when answering the evaluation questions. The Evaluation Team must review situational analysis and the current list of 17 major crime indicators (broken down by major crimes groups) on citizen security that have been developed and improved for the Northern Triangle Countries as a result of InfoSegura actions, and how efforts to improve their analysis and comparison between countries are connected to the program results. Key informant interviews, surveys, focus group discussions The Evaluation Team will interview stakeholders, through key informant interviews, group interviews, short surveys, and/or focus groups discussions. The Evaluation Team will include both men and women in the stakeholders’ consultation processes. USAID/CAM and/or UNDP will provide key informant contact information once the evaluation begins. At minimum, the Evaluation Team will interview: • Key USAID/CAM staff (Agreement Officer’s Representative, AOR, of the InfoSegura activity, alternate AOR of the InfoSegura activity, RDCS Development Objective 3 Team Leader, Democracy and Governance Office Director, among others) and Key USAID/Guatemala and USAID/Honduras staff (Offices Directors and/or bilateral activity managers) • InfoSegura staff (InfoSegura Regional Leadership based in San Salvador and individual InfoSegura national teams in El Salvador, Guatemala, and Honduras) • InfoSegura key partners: Central America Central and Local Governments40 o El Salvador: Ministry of Justice and Security, General Directorate for the Social Prevention of Violence and Peace Culture (PREPAZ), Directorate of Information and 40 InfoSegura only provides support to national level institutions (ministries, secretariats, etc.). Those institutions are the ones who reach the local governments who are indirect beneficiaries of InfoSegura. 37 Analysis (DIA), General Directorate of Statistics and Censuses (DIGESTYC), General Directorate of Prisons, National Civilian Police (PNC), National Council on Citizen Security, Secretariat for Governance, Ministry of Education (MINED), Technical Secretariat of the Presidency (SETEPLAN). o Honduras: Ministry of Security and its analytical and citizen security policymaking departments, Technical Unit for Institutional Coordination (UTECI), Secretariat of the Presidency, National Information Center for Social Sectors (CENISS), Studies and Analysis Center for Citizen Security (CEASCI), Local Government Support Unit (UGAGLO), municipal/department observatories selected in coordination with USAID/Honduras “Juntos en Acción por la Convivencia” activity and the Government of Honduras. o Guatemala: Ministry of Governance and its analytical and statistical department, Vice Ministry of Violence and Crime Prevention who coordinates the “Mesa Técnica Interinstitucional,” National Statistics Institute of Guatemala (INE), National Police, Unidad Para la Prevención Comunitaria de la Violencia (UPCV), select departmental observatories. • InfoSegura key partners: Civil Society Organizations/Academia o El Salvador: “Dr. Guillermo Manuel Ungo” Foundation (FUNDAUNGO), Cristosal, Organization of Salvadoran Women for Peace (ORMUSA), “Norma Virginia Guirola de Herrera” Institute of Women’s Studies (CEMUJER), Latin American School of Social Sciences (FLACSO, by its acronym in Spanish), and “José Simeón Cañas” Central American University. o Honduras: National Autonomous University of Honduras and its research wing on crime, the Association for a More Just Society, and FLACSO/Honduras. o Guatemala: Rafael Landívar University, FLACSO/Guatemala. • UN Country Teams and Agencies: UNDP, International Organization for Migration (IOM), United Nations High Commissioner for Refugees (UNHCR), United Nations Entity for Gender Equality and the Empowerment of Women (UNWOMEN), United Nations Population Fund (UNFPA), United Nations Office on Drugs and Crime (UNODC). • Other USAID Implementing Partners working in the citizen security field in the region through the implementation of regional activities (IOM, Pan-American Development Foundation, and the John Jay University of Criminal Justice) or bilateral activities (Implementing Partners to be recommended by USAID/Guatemala and USAID/Honduras) The Evaluation Team is encouraged to interview other bilateral donors and key stakeholders as needed involved with Citizen Security cooperation programs (namely Germany and Spain); multilateral donor partners such as the World Bank and the Inter-American Development Bank; and other actors that can provide an insight into USAID programmatic impacts. 38 A sampling plan describing the selection process (such as purposeful, random, or a combination of approaches) 41 for organizations and stakeholders for key informant interviews, surveys, and focus groups discussions (including sex disaggregation) is expected to be included in the Evaluation Plan and Final Report. Site visits and direct observation In consultation with USAID/CAM and UNDP, the Evaluation Team will select relevant site visits based on a sampling plan developed for the Evaluation Plan and included in the Final Report. At minimum, the Evaluation Team should expect to visit the capital cities of each one of the three Northern Triangle countries of Central America where the Central Government institutions are located. In addition, when feasible and with the support of UNDP, the host country government, and USAID in each country; the Evaluation Team should identify and visit municipalities42 where the assisted institutions have applied the knowledge transferred through InfoSegura. Some examples of municipalities of interest to USAID are: Guatemala Metropolitan area of Guatemala (Guatemala City, Villa Nueva, Mixco) and Chimaltenango (53 miles from Guatemala City – asphalt road and dirt road) Honduras Tegucigalpa and San Pedro Sula (157 miles from Tegucigalpa – mostly asphalt highway to varying degree of quality) El Salvador San Salvador, Zacatecoluca (38 miles from San Salvador – high quality asphalt highway), and Colón (11 miles from San Salvador – high quality asphalt highway) The Evaluation Team should plan to meet with USAID bilateral missions (USAID/Honduras and USAID/Guatemala) after being introduced by USAID/CAM to get feedback in the field since the Offices of Democracy and Governance (Guatemala and El Salvador) or Program Office/Democracy and Governance Office (Honduras) monitor and coordinate InfoSegura implementation with Mission efforts. When meeting key in-country representatives, the Evaluation Team may be accompanied by a member of USAID bilateral missions if Missions consider it appropriate. The Evaluation Team will be required to debrief USAID bilateral missions on field meetings or site visits, if deemed appropriate. The purpose of the debriefing will be to share findings and receive comments or feedback prior to the preparation of the Draft Evaluation Report. The Evaluation Team may attend events hosted or sponsored by InfoSegura during the fieldwork period of the evaluation to conduct direct observation. The Evaluation Team can use these events to talk with stakeholders, conduct interviews and collect additional data as evidence to answer 41 Some sampling approaches include quota sampling, proximity sampling, convenience sampling, theoretical sampling, typical case sampling, etc. 42 National institutions assisted by InfoSegura have supported the local governments; therefore, it is expected that stakeholders at the local level who are indirect beneficiaries are not aware of the work of InfoSegura. 39 the evaluation questions. USAID/CAM or UNDP will provide the Evaluation Team with a list of events once the evaluation begins. Team planning meetings An initial team-planning/kick off meeting will be held in El Salvador between USAID/CAM and the Evaluation Team before the submission of the Evaluation Plan so that USAID/CAM can clarify any questions from the Evaluation Team, expectations, and guidelines. The expected results of this meeting are to: • Clarify each team member’s role and responsibilities; • Confirm the anticipated timeline and deliverables; • Discuss data collection tools and methodologies by evaluation question to be presented in the Evaluation Plan; and • Identify communications logistics and how the Evaluation Team, USAID/CAM, and UNDP will communicate with each other. Additional meetings may be held as deemed necessary by USAID/CAM and/or the Evaluation Team. E. Deliverables It is estimated that not more than 70 working days of services from the starting date of the evaluation will be required to complete a high-quality evaluation as required under this SOW. During that timeframe, the evaluation team shall submit the following deliverables: 1. An Evaluation Plan, in Word Gill Sans font43 size 12, to be completed by the Evaluation Team after the Team Planning Meetings, no later than 15 calendar days after the starting day of the evaluation. USAID/CAM will receive the Evaluation Plan via electronic mail and review it to provide comments no later than 10 working days after receiving the document. The Evaluation Plan will provide details of how the various deliverables, tasks, and activities will be undertaken. It must include at least: • InfoSegura description and logic (theory of change/development hypothesis); • Evaluation design, 44 and the explanation of why one design or mix of designs proposed is the most appropriate, its limitations, and how these limitations will be addressed; • A matrix summarizing the following information per each evaluation question: o Method(s) for data collection, data source, the explanation of why one method or mix of methods is the most appropriate, its limitations, and the ways to address them; and 43 If the Evaluation Team does not have Gill Sans family available, they must use any other approved font as per the USAID Graphics Manual and Partners Co-Branding Guide. 44 Some examples of evaluation designs for performance evaluations include snapshot design, cross-sectorial design, before-and-after design, time series design, case study design, panel design, etc. 40 o Technique(s) for data analysis, 45 the explanation of why one analysis technique or mix of techniques is the most appropriate, its limitations and the ways to address them. • Data Management Plan describing the capture of data (for example, interview notes or live recording), storage and transfer, and how all data will be handled in such a manner as to protect the identities of informants in any situations where their comments could potentially have a negative impact on their employment or security. • Timeline and/or Milestone Plan, including tentative starting time for data collection and duration of each activity conducted under the evaluation; • Drafts of data collection protocols, such as questionnaires or focus group moderator guide(s), interview scripts, consent form,46 etc.; • Evaluation Team composition and roles; • Location for the evaluation and Site visit plan. If the Evaluation Plan includes key informant interviews, surveys, and/or focus group discussions, the Evaluation Plan should include the following information: o How the interviews/surveys will help to answer the evaluation questions; o Who will conduct the interviews/surveys and why they are qualified to do so; o What the rationale and methods are for deciding the number, timing, and location of the interviews/surveys; o How the participants will be selected and recruited; o How the interviews/surveys will be recorded; and o How the interview/survey data will be analyzed and presented. The Evaluation Plan, particularly the data collection and analysis protocols, as well as interview and focus group guides must be approved by USAID/CAM prior to the start of data collection and the field work. All interview protocols must be submitted in English and Spanish. The Evaluation Team will have another five (5) working days to make any changes. Once the Evaluation Plan is approved, the Evaluation Team will submit an electronic copy of it in PDF format to USAID/CAM. Any subsequent change to the Evaluation Plan must be approved by USAID/CAM. The Evaluation Team shall provide USAID/CAM and UNDP with a preliminary briefing on the Evaluation Plan prior to the beginning of data collection. 2. Brief weekly bullet reports of activities, submitted by email to the manager of this evaluation every Monday by the close of business. 3. A Preliminary Findings Briefing for USAID/CAM (USAID/Guatemala and USAID/Honduras can join the briefing via video teleconference system), UNDP, and other stakeholders that USAID/CAM consider necessary on the preliminary findings identified by the Evaluation Team immediately after finalizing the data collection phase 45 Some examples of data analysis techniques include parallel, conversion, sequential, multilevel, data synthesis, content analysis, etc. 46 If underage persons (less than 18 years old) will participate in this performance evaluation, the Evaluation Team must make sure to comply with all national regulations related to Child Protection. 41 and before starting the draft report. According to the audience, the Draft Report briefing may be conducted in English or Spanish. Only the Team Leader needs to be present for this briefing; however, local/regional Evaluation Team members may also attend. The Evaluation Team will use the Preliminary Findings Briefing as a feedback exercise to prepare the Draft of the Final Report. 4. A Draft of the Final Report in Word, Gill Sans font size 12, submitted for review due no later than 60 calendar days after the approval of the Evaluation Plan via electronic mail. USAID/CAM will be responsible for distributing it to the implementing partner and other stakeholders for comments. USAID/CAM will consolidate all comments and send the draft back to the Evaluation Team within 10 working days. At a minimum, and in accordance with the USAID Evaluation Policy and Automated Directives System (ADS 201), the Final Report and its draft versions must include the following sections: • Executive Summary of the purpose, background, evaluation questions, findings, conclusions, and recommendations; • Evaluation purpose and questions; • Thorough description of the evaluation design and any challenge/limitations, 47 with emphasis on the timeliness and methods for data collection and data analysis; • Relevant data analysis tables; • Findings and conclusions drawn from the analysis of the findings; 48 • Action-oriented, practical, and specific recommendations with defined responsibility for the action; • A dissemination plan of findings, conclusions, and recommendations to intended users of the evaluation; and • Appendices: o Original SOW, annotated with any changes approved by USAID; o Evaluation and data collection team composition and roles, with conflict of interest disclosures for all real or perceived conflicts of interest, if necessary; o Data collection protocols and instruments including questionnaires and checklists; o Review matrix of documents consulted; o Meeting notes; o Complete schedule of evaluation activities, meetings, and interviews; o List of individuals and organizations contacted and sites visited; and o Tables, graphs, pictures taken during site visits, maps. USAID/CAM expects to receive a high-quality Draft Report from the Evaluation Team. USAID will assess the quality of it using the Evaluation Report and Review Template. 47 The Evaluation Team must identify (a) steps taken to mitigate limitations, and (b) how/whether the limitations affect any particular finding, conclusions, or recommendations. 48 In moving from findings to conclusions, the analysis must be clear as to how findings are synthesized through different techniques such as divergence, convergence, and amalgamation; propensity; weighting; etc. 42 5. Final Report in PDF, font Gill Sans size 12, no longer than 40 pages in its body, excluding the cover page; Table of Contents; List of Acronyms; and Appendices. The approved Final Report must adhere to USAID’s Evaluation Policy and ADS 201, Criteria to Ensure the Quality of the Evaluation Report, and must be submitted in English and Spanish and have incorporated USAID’s comments, as appropriate. The Final Report will be due to USAID 10 working days after the Evaluation Team receives comments on the draft. Five high￾quality printed, bound copies in English and Spanish of the Final Report must be submitted to USAID within 10 calendar days of acceptance of the Final Report. 6. A One-Page summary of the evaluation purpose, findings, conclusions, and recommendations. The One-Page summary will be prepared in English and Spanish in PDF. 7. Any raw data (qualitative or quantitative) collected in electronic form (DVD or flash drive, in original format of Word, Excel, etc.) is due no later than 100 calendar days after the starting date of the evaluation. As per ADS 579, the Evaluation Team must submit to the Development Data Library (DDL), in a machine-readable, non-proprietary format, a copy of any datasets that are used (or of sufficient quality) to produce an Intellectual Work. 8. One event to present the content of the Final Report for key Central Government representatives and other stakeholders in each of the three countries of the Northern Triangle (Guatemala, Honduras, and El Salvador). 9. Other deliverables as identified during the Team Meeting and agreed to by USAID and the Evaluation Team. All reports and papers will be considered draft versions until they are approved by USAID. These draft documents must be labeled with the word “DRAFT” in watermark. Findings must be presented as analyzed facts, strong qualitative and quantitative evidence and data, and not based on anecdotes, hearsay or the compilation of people’s opinion. To ensure unbiased findings, there is no guarantee that findings will be modified based on USAID suggestions. The Evaluation Team will research, investigate, and corroborate as objective any suggestion before it is incorporated in the findings, and the change will be noted in the draft document so as to have a record of the change. All submitted reports and presentations must be thoughtful, well-researched, and well-organized documents, and objectively answer the evaluation questions. When writing the report, the Evaluation Team must remember the different audiences. The style of writing should be easy to understand and concise, while making sure to address the evaluation questions and issues with accurate and data-driven findings, justifiable conclusions and practical recommendations.49 The Evaluation Team should clearly list any biases or limitations that exist during both data collection and analysis (selection bias, recall bias, unobservable differences between comparator groups, etc.). In addition, all real or possible conflicts of interest must be disclosed by each member of the Evaluation Team in writing. 49 For additional information on the criteria to ensure the quality of Evaluation Reports, see the USAID’s How-To Note “Preparing Evaluation Reports”, and Annex 1 of the USAID Evaluation Policy. 43 When quoting an individual in any report, the Evaluation Team must always give the context or circumstances of the quote. Correcting a grammatical error in the quote may be valid, but not rewording an entire phrase. When translating quotes from one language to another, the Evaluation Team should do so in an idiomatic way and care must be taken to ensure that the tone of the translation is equivalent to the tone of the original. Quotes should be presented in their original language in report texts. All reports must comply with the USAID Graphic Standards Manual and the ADS Style and Format Guide. Once a Final Report has been approved by USAID, the Evaluation Team will make it compliant and submit it to USAID’s Development Experience Clearinghouse (DEC).50 The Evaluation Team will email USAID the DEC link where the evaluation reports. USAID/CAM may attach a Statement of Differences as an Annex to any Final Report if any differences remain in the final version. F. Evaluation Management Evaluation Team This performance evaluation will use a combination of multidisciplinary international, regional, and local experts. The Evaluation Team must include at a minimum the following three positions: Evaluation Team Leader Minimum qualifications Education: Master’s degree in fields such as Criminal Justice, Criminology, International Crime and Justice, Crime Prevention, Political Science, or Public Administration (with a focus on criminal justice or crime prevention), Ph.D. or doctorate degree or professional with Doctoral candidacy is a plus. Formal training in monitoring and evaluation is a plus. Language Proficiency: American English Level IV and Spanish Level III Work Experience: At least eight years of relevant prior experience conducting rigorous external evaluations using both quantitative and qualitative methods for development objectives and monitoring projects and programs overseas, preferably in Latin America. Experience in Central America is a plus. At least eight years of project management experience in development is required. Experience with management of multidisciplinary teams is a plus. Familiarity with USAID’s objectives, approaches, operations, and policies, particularly as they relate to evaluations is a plus. Role: The Evaluation Team Leader will be responsible for overseeing and coordinating all activities related to this performance evaluation and for ensuring the production and completion of quality deliverables in a professional manner, in conformance with this SOW. 50 Per ADS 540, documents and development assistance projects materials produced or funded by USAID must be submitted for inclusion in the DEC. 44 Data and Analytics Specialist Minimum qualifications Education: Master’s degree in fields such as Economics, Statistics, Engineering, Social Research, or other fields related to Data Analytics. Formal training in monitoring and evaluation is preferred. Language Proficiency: Spanish Level IV and American English Level IV. Work Experience: At least five years of progressively responsible, professional-level experience in statistics, data analytics, and public administration and policy with a focus/specialization in crime prevention/criminal justice is a plus, with prior experience preferably in Central or Latin America. Familiarity with USAID’s objectives, approaches, and operations, particularly as they relate to evaluations is a plus. Role: The Specialists will provide advice and analysis to the Evaluation Team Leader and will lead work on evaluating InfoSegura’s work on data collection, comparison, analysis, and communication. He/she should be highly experienced with statistics and data analytics, as they relate to public policy and criminal justice. Prior experience working on the issues in Central or Latin America is a plus. Governance Specialist Minimum qualifications Education: Master’s degree in fields such as Political Science, Law, Criminal Justice, International Crime and Justice, or Public Administration (with a focus on criminal justice or crime prevention and/or governance), with relations to development and/or public policy is required. Language Proficiency: Spanish Level III and American English Level IV. Work Experience: At least five years of progressively responsible, professional-level experience conducting evaluations related to public administration and policy, or good governance is required, preferably in Central or Latin America. Experience or knowledge of Central America’s major governance challenges, particularly crime and violence issue are preferred. Some familiarity with USAID’s objectives, approaches, and operations, particularly as they relate to evaluations is a plus. Role: The Specialist will provide advice and analysis to the Evaluation Team Leader and will assist the Evaluation Team Leader and the Data and Analytics Specialist with work on evaluating InfoSegura’s work on public policy design, institutional development, governmental transparency and government-civil society collaboration on citizen security. He/she should be familiar with governance and public policy models as they relate to international development, along with governance and citizen security issues in Central America. The Evaluation Team should have considerable experience in designing, monitoring, and evaluating development assistance programs. They must have proficiency in MS Office Suite and have excellent written, proofreading, and oral presentation skills and the ability to conceptualize and write clearly and concisely and attention to details. Understanding of the Latin American context is necessary, with a preference for personnel with work experience specifically in Central America. At least one key personnel member must have experience working in gender and social inclusion issues in a development context, transforming qualitative data, analyzing quantitative data, and producing data visualization in an easily digestible format. 45 All Team members will be required to provide in advance to USAID a signed statement indicating any conflict of interest, real or in appearance. The Team Leader must be external to USAID. No Evaluation Team member shall have been directly involved in the design and implementation of InfoSegura in any stage. Anyone who has directly employed by USAID or UNDP in the last five years must not be considered as part of the Evaluation Team. Logistics The Evaluator will be responsible for all logistics support under this SOW, including field office administration, all travel arrangements (with required USAID clearances), team planning facilitation and appointment scheduling, coordination with all partners and stakeholders involved, administrative services (computer support, printing and copying), report editing and dissemination, and for complying with provisions set forth in this SOW. USAID/CAM will provide limited support to the Evaluation Team. This support, if needed, may include assistance in arranging high-level meetings; access to the U.S. Embassies compounds as necessary; and access to all reports, data, and other relevant documents created by InfoSegura. USAID representatives may accompany the Evaluation Team for some or all the evaluation, especially in high level meetings. The Evaluation Team is expected to consider this when making logistical arrangements. 46 ANNEX II: COUNTRY CONTEXT FOR EL SALVADOR, GUATEMALA, AND HONDURAS El Salvador Between 1980 and 1982, El Salvador experienced a civil war that killed 70,000 people and displaced over a million.51 The conflict ended with the 1992 peace accords between the El Salvador government and the insurgent Farabundo Martí National Liberation Front (FMLN). Subsequently, the leftist FMLN became a political party. Over the post-war years, the country has lapsed into serious criminality, with at least 93,000 murders since 1993, more than half of them attributed to two gangs, Mara Salvatrucha (MS-13) and the 18th Street gang. Their active members and support bases are thought to include eight percent of the country’s 6.2 million people. This gang-related violence has led to a security crisis, which successive governments by turn have struggled unsuccessfully to address. The country’s main public-security bodies include the two political parties, FMLN and the Nationalist Republican Alliance (ARENA), a judicial system with an unequal resource distribution, and a military-backed police force. The country’s 14 departments and 262 municipalities depend mostly on the central government for the design and implementation of security policy. Security powers fall largely to the Ministry of Justice and Public Security, which operates the police and the prisons. Both FMLN and ARENA have to date followed a largely punitive approach. Charged with investigating and trying crimes are the prosecutor’s office, the police, and the judiciary. Extreme violence and legal prosecution cum mass detention have impaired the courts and led to vastly overcrowded prisons. The police have come under mounting pressure to combat crime and gang attacks. Most of the 28,000 officers—facing active gang members reckoned at 60,000—are of lower-class origin, have low salaries, and thus have to live in gang-controlled areas, exposing them and their families to substantial risk.52 Instead of community policing, the police engage in armed raids and often firefights. This has led to deeper involvement of the military— the institution with the highest public approval rating—in public security. Corruption in judicial and security entities compounds this situation. The latest effort to confront the violence problem is a project called “Safe El Salvador” under the Sánchez Cerén government. Although the focus of security policy has continued to favor law enforcement, the project initiated some violence prevention measures. The project, costing about (US$)200 million per year and financed mostly by donor funds and to a lesser extent by special Salvadoran taxes, began by working in municipalities most affected by violence.53 UNDP’s implementation of InfoSegura has faced numerous challenges beyond government personnel changes (linked to political change) throughout the implementation period. When InfoSegura began in 2014, these challenges included:54 1) weak institutional and technical capacity; 2) national-level public officials did not prioritize institutional strengthening; 3) lack of credibility in statistics used to take policy decisions; 4) lack of clear guidelines for gathering, analyzing, and using data; 5) high levels of mistrust and poor coordination in public institutions and civil-society; 51 Most of this section on El Salvador relies on information taken from “International Crisis Group,” Dec. 19, 2017. 52 The MS-13 gang is estimated to earn $31.2 million per year, mostly through extortion. 53 Of the $93 million collected in taxes in 2017, around 70 percent went to finance the police and the armed forces. 54 PNUD, Avance de implementación del proyecto InfoSegura. 10 de octubre de 2018 47 and 6) limited capacity of civil-society entities to educate and conduct research. Honduras Honduras is a low middle-income country that faces major challenges. According to official data, more than 66 percent of the population lived in poverty in 2016. In rural areas, about one in five Hondurans live in extreme poverty. The country’s economy grew by 4.8 percent in 2017, according to the latest estimates; and a 3.6 percent growth is expected for 2018. Calculations show that over the 1960-2017 period, on average the per-capita income grew only 1.2 percent.55 The country faces the highest level of economic inequality in Latin America. Honduras’ persistent poverty is the result of long-term low per-capita growth and high income inequality, worsened by the country’s high vulnerability to shocks. Rampant crime and violence are other major challenges.56 Honduras’ history of low and volatile economic growth and high income inequality has created the conditions for the emergence of two mutually reinforcing cycles: (a) a high crime-low growth cycle; and (b) a high emigration/remittance flows-low growth cycle. For the past 15 years, these cycles have shaped the country’s challenges and have continued to impact its growth prospects.57 Honduras has one of the highest levels of crime and violence in the world (43.6 murders per 100,000 inhabitants in 2017, according to the Observatory of Violence at the UNAH),58 undermining growth and stifling economic opportunity. Honduras has been strategically important to the drug trade since the 1970s, but in the last 15 years the country has witnessed an increase in gang activity and a rise in violent crimes like extortion and kidnapping. Young men are particularly vulnerable to crime, both as victims and perpetrators. The cost of crime in Honduras is estimated to have reached a staggering 10 percent of gross domestic product (GDP) (US$900 million) per year, and the health costs alone are estimated at 1.3 percent of GDP.59 Low economic growth has a negative impact on the creation of economic opportunities and this, in turn, lowers the opportunity cost of joining criminal groups. About 13 percent of the country’s working-age population live in the U.S. Remittances approached 18 percent of GDP in 2014, the largest rate in Central America and among the top 15 worldwide. The average monthly remittance reported by recipients is (US$)477, almost 2.5 times the country’s average monthly per capita income. Migration opens new possibilities for workers, and remittances have contributed to poverty reduction. Yet evidence from Honduras indicates that large emigration and remittances flows have also negatively impacted productivity by shrinking the size of the labor force and increasing reservation wages. Low institutional quality is a root cause behind the country’s development outcomes. For instance, Honduras scores in the lowest quartile in the world in the areas of control of corruption, rule of law, and government effectiveness; and scores below average in the areas of 55 The World Bank in Honduras. 2018. Data. API_HND_DS2_en_excel_10314893.xls. https://data.worldbank.org/country/honduras 56 The World Bank in Honduras. 2018. Overview. http://www.worldbank.org/en/country/honduras/overview 57 United Nations Office on Drugs and Crime (UNODC). 2014. Global Study on Homicide 2013 (United Nations publication, Sales No. 14.IV.1). https://www.unodc.org/documents/gsh/pdfs/2014_GLOBAL_HOMICIDE_BOOK_web.pdf 58 The World Bank in Honduras. 2018. Overview. http://www.worldbank.org/en/country/honduras/overview 59 World Bank. 2011a. Crime and Violence in Central America: A Development Challenge—Main Report. Washington, DC: World Bank. https:// openknowledge.worldbank.org/handle/10986/2744. 48 political stability, voice and accountability, and regulatory quality. The persistence of crime is attributable at least in part to the judicial system’s lack of enforcement capacity. A self-assessment in Chapter III of Criminalization and Law Enforcement of the United Nations Convention on Anti-Corruption revealed that major challenges remain in terms of law enforcement, with few cases being investigated, prosecuted, and eventually punished with convictions.60 Guatemala The 1996 Peace Accords that brought to a close a 36-year conflict sought to promote an ambitious and far-reaching agenda that aimed to address the structural causes of the conflict. However, advances in equity and social inclusion have been limited, and Guatemala continues to face a series of governance challenges due to institutional fragility, impunity, and high levels of criminality and violence. Guatemala has shown strong economic growth in recent years.61 Yet the World Bank notes that the country “has one of the highest inequality rates in Latin America, with some of the worst poverty, malnutrition and maternal-child mortality rates in the region, especially in rural and indigenous areas.”62 Despite the fact that Guatemala reduced its poverty rate to 51 percent between 2000 and 2006, according to the World Bank study, “Poverty Assessment in Guatemala,” official figures indicate that poverty rose to 59.3 percent in 2014. Fifty-two (52) percent of all people living in poverty are indigenous.63 The World Bank further notes that “High levels of crime and violence represent staggering economic costs for the country.”64 Guatemala’s murder rate is one of the highest in the world in 2008, with 48 murders per 100,000 people. Authorities attribute as much as 45 percent of the homicides to drug trafficking activities. As in the other countries of Central America, strengthening institutional capacities in the area of citizen security has been a constant objective of numerous national and international development projects since the signing of Peace Accords in 1996. The international donor community has provided significant financial and human resources to improve national decision-making processes and capacities for strategic, tactical, and operational planning to combat and prevent crime and violence. Specific efforts have been made to improve the production of relevant statistical information and its application by different citizen-security institutions. Such efforts face diverse challenges, among them the broader volatility due to political crises. Throughout 2018, there were a series of significant changes at the management level in the MINGOB and rotation in some of the police directorships. However, a positive note in the context of political and institutional uncertainties is that the technical staff of the Vice Ministry 60 Hernandez Ore, Marco Antonio, Liliana D. Sousa, and J. Humberto Lopez. 2016. “Honduras: Unlocking Economic Potential for Greater Opportunities.” Systematic Country Diagnostic. World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO. Downloaded from http://documents.worldbank.org/curated/en/519801468196163960/pdf/103239-v2-PUB-P151906-Box394858B￾PUBLIC-DOI-10-1596K8570-EPI-K8570.pdf 61 “Latin America in recent years, with a GDP growth rate of 3.0 percent since 2012 and 4.1 percent in 2015. In 2017, the country’s economy grew by 2.8 percent, according to the latest estimates, and is expected to grow by 2.6 percent in 2018.” 62 World Bank. 2018. “The World Bank in Guatemala.”https://www.worldbank.org/en/country/guatemala/overview 63 World Bank. 2009. “Guatemala Poverty Assessment.” 64 Ibid. 49 of Violence and Crime Prevention, InfoSegura’s counterpart, seems to enjoy a relatively high degree of stability. 50 ANNEX III: DATA COLLECTION INSTRUMENTS Protocols for Questions from SOW Protocol – EQ 1 • To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? o To what extent has InfoSegura developed the capacity within the Sub-Technical Units (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? o To what extent has InfoSegura developed the capacity of source institutions (such as the National Civilian Police, the Attorney General’s Office, and the Forensic Medicine Institute in El Salvador) to ensure quality of data and improve information sharing between source institutions and Sub-Technical Units? o What is the potential that the capacity developed within the Sub-Technical Units and the source institutions increases or decreases after the completion of InfoSegura? • You are familiar with InfoSegura. We would like to ask you the following questions to learn your opinion: (Note: This question is very open and is intended to give the respondent ample latitude of response.) o How has InfoSegura led to changes in the quality of crime and violence data? Please explain. o And has it led to collection of data in a more-timely manner? o And how has it affected the recording of crime and violence data? How and where is the data recorded? o How could InfoSegura be improved? • This question has three parts: In what measure has InfoSegura improved the capacity within your unit to: 1) collect, 2) record, and 3) analyze crime and violence data? How could InfoSegura be improved? • Your agency collects crime and violence data. This question has two parts: 1) In what measure has InfoSegura improved the quality/reliability of the data that you collect? 2) In what measure has InfoSegura improved information sharing between your agency and sub-technical units in the Ministry of Security? How could InfoSegura be improved? • Think about how InfoSegura has affected the capacity within your unit. Is the work of InfoSegura sustainable—i.e., will the capacity remain after InfoSegura support ends? How could the work of InfoSegura be improved to increase the chances of sustainability? • Do you have any further recommendations for InfoSegura’s design or its management? 51 Protocol – EQ 2 • To what extent have Central American governments used (as planned) crime and violence data—produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions? o What are the internal and external factors that foster or impede the Governments? In what measure has the government of El Salvador/Honduras/Guatemala used crime and violence data that InfoSegura and other entities have produced under InfoSegura in order to make citizen-security policy decisions? • This question has two parts: What factors have fostered government use of the data? What factors have impeded government use of the data? • Do you have any further recommendations for InfoSegura’s design or its management? Protocol – EQ 3 • To what extent has the creation of partnerships and transparency (CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy? o What are the internal and external factors that foster or impede the collaboration among the members of CONOSE? • This question has two parts: Has InfoSegura fostered 1) partnerships and 2) transparency between government and civil society? How has this happened? • Have these partnerships and transparency led to changes in 1) national collaboration on designing citizen security policy? 2) In regional collaboration on the design of citizen-security policy? What kinds of changes? • What factors 1) foster collaboration among members of CONOSE? 2) What factors impede collaboration among members? • What can be done to improve collaboration between 1) national governments and civil society? And 2) between regional actors? • Do you have any further recommendations for InfoSegura’s design or its management? 52 ANNEX IV: INFOSEGURA THEORY OF CHANGE HYPOTHESIS AND RESULTS FRAMEWORK The results framework originally prepared by USAID is shown in blue boxes below the broken line in Figure 1. This refers to InfoSegura itself. But as the evaluation SOW notes, InfoSegura also supports other USAID projects, indicated in green boxes above the dotted line in Figure 1, while change flows upward along the lines and arrows. The UNDP’s results framework is shown in Figure 2. Since UNDP has designed and implemented InfoSegura in its entirety, its framework is more important to the ET. Figure 1: InfoSegura Change Hypothesis Figure 1: InfoSegura Change Hypothesis DO3 Regional human rights and citizen security improved (CAM Strategy) IR 3.1 Regional capacity to address citizen security improved through more coordinated governance systems (IR 3.1, CAM Strategy) Component 1: Evidence-based policy making on citizen security strengthened Component 2: Regional collaboration and networking on citizen security increased 2.1 Dialogue on citizen security issues between civil society and government institutions improved Public Safety Improved and Access to Legal System Enhanced (A4P Action Line) 1.1 Quality and comparable citizen-security indicators developed 1.2 Analysis capacity to inform citizen-security policy decisions strengthened 2.2 Exchange of citizen security best practices and effective strategies increased Security enhanced and Levels of Crime and Violence Reduced (CEN Strategy, Goal 2, Pillar 2) ?? InfoSegura links to these projects Purpose of InfoSegura: To strengthen evidence-based policy making, and increase regional coordination and collaboration of effective citizen security strategies 53 Figure 2: InfoSegura Results Framework The two components in Figure 3 show the results from USAID’s Results Framework shown in Figure 2. The boxed descriptions below each component are activities that achieve these two results, thus the upward-pointing arrow. Note that the UNDP also worked in countries outside of the Northern Triangle (e.g., Dominican Republic, Panama, Belize, and Costa Rica); hence, the mention of those countries in the descriptions in Figure 3. The reader should look only at the Northern Triangle countries. 54 ANNEX V: EVALUATION MATRIX Evaluation Question Data Sources Data Collection Method Data Analysis Method 1. To what extent has InfoSegura contributed to change the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? To what extent has InfoSegura developed the capacity within the STUs (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? To what extent has InfoSegura developed the capacity of source institutions (such as the PNC, Attorney General’s Office, and IML in El Salvador) to ensure quality of data and improve information sharing between source institutions and STUs? What is the potential that the capacity developed within STUs and the source institutions increases or decreases after the completion of InfoSegura? • Documentation • Monitoring data on performance indicators • Current list of 17 major crime indicators on citizen security • Available survey data • Key USAID/CAM staff • InfoSegura regional leadership based in El Salvador and individual InfoSegura national teams in El Salvador, Guatemala, and Honduras. • InfoSegura key partners in El Salvador: PREPAZ, DIA, DIGESTYC, General Directorate of Prisons, PNC, CNSCC, Secretariat for Governance, MINED, SETEPLAN. • InfoSegura key partners in Guatemala: MINGOB, analysis and statistics department within the MINGOB, MTI, INE, PNC, UPCV, Municipal Observatories. • InfoSegura key partners in Honduras: Ministry of Security, analysis and citizen security policymaking department within the Ministry of Security, UTESI, CENISS, CEASCI, UGAGLO, Municipal Observatories to be selected in coordination with USAID Honduras, “Juntos en Acción por la Convivncia,” and Government of Honduras (GOH). • InfoSegura key partners - Civil Society Organizations/Academy: El Salvador (FUNDAUNGO, FLACSO, UCA, Cristosal, ORMUSA), Guatemala (Rafael Landívar University, FLACSO), Honduras (Violence Observatory of UNAH, FLACSO). • Relevant UN Country Teams and Agencies: IOM, UNHCR, UNFPA, UNODC. • Other relevant USAID implementing partners to be selected in coordination with USAID. IIs conducted on a convenience sample of key informants selected from the key partners of InfoSegura. • Content analysis • Descriptive analysis of response obtained from the sample of key informants. • Concordance analysis of data obtained from the IROs and data obtained from the data applications in web sites. 55 Evaluation Question Data Sources Data Collection Method Data Analysis Method 2. To what extent have Central American governments used (as planned) crime and violence data produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security￾related policy decisions? • Ministry of Security in El Salvador, Honduras, and Guatemala • UNDP in El Salvador, Honduras, and Guatemala • Attorney General’s office in El Salvador, Honduras, and Guatemala • Desk review of performance data and goal outcome measures • KIIs Content analysis -What are the internal and external factors that foster or impede the Governments to use the data? • Ministry of Security in El Salvador, Honduras, and Guatemala • UNDP in El Salvador, Honduras, and Guatemala • Selected donors KIIs Content analysis 3. To what extent has the creation of partnerships and transparency (CONOSE) between government and civil society led to changes on national and regional collaboration on the design of citizen security policy? • CONOSE in El Salvador, Honduras, and Guatemala • UNDP in El Salvador, Honduras, and Guatemala • Desk review of performance data and goal outcome measures • KIIs • Group discussions • Survey of CONOSE stakeholders Content analysis -What are the internal and external factors that foster or impede the collaboration among the members of CONOSE? • CONOSE in El Salvador, Honduras, and Guatemala • UNDP in El Salvador, Honduras, and Guatemala • KIIs • Group discussions • Survey of CONOSE stakeholders Content analysis 56 ANNEX VI: SOURCES OF INFORMATION PERSONS INTERVIEWED Name Position Organization El Salvador Aparicio, José Ignacio Data-Quality Support Specialist, Quality Control Team MEL Initiative Amaya, Edgardo Vice Minister, Information and Analysis DIA/MJS Barahona, Erick Sub-Director of Information and Analysis National Statistical Data Directorate of Information and Analysis Carsana, Daniel Coordinator, InfoSegura El Salvador UNDP/InfoSegura De Valencia, Annie Advisor, Gender and Social Inclusion USAID Gordillo, Juan Pablo InfoSegura Deputy Regional Project Coordinator UNDP Guzmán, Silvia Program Analyst UNDP Hernández, Evaristo Director DIGESTYC Hidalgo, Orlando Monitoring, Evaluation, and Learning Specialist, Program Office USAID Hublin, Trevor Agreement Officer’s Representative (AOR), InfoSegura USAID/El Salvador/Region Jaime, Doris Director of Monitoring and Evaluation SETEPLAN López, Raúl Vice Minister MJS McKenney, Robert Team Leader for Citizen Security, Office of Democracy and Governance USAID/El Salvador Merino Domínguez, Mónica Deputy Representative UNDP Morales, José Antonio Sub-Secretary of Governance Secretariat of Governance Oliva, Andreu, S.J. Rector UCA Palencia, Guadalupe M&E and Learning Specialist, Democracy and Governance USAID Paz, Luis Inspector DIGESTYC Ramírez, Hugo Armando Sub-Director of Public Security PNC Ramírez, Pablo Technical Advisor, Sub-Directorate of Public Security PNC Ramos, Carlos Professor, College of Social Science and Member of Executive Committee CONOSE Saud, Raúl Analyst DIGESTYC Schaffer, Chris Director, Office of Democracy and Governance USAID El Salvador/Region Smut, Marcela Regional Coordinator and Director InfoSegura UNDP Swarin, Gregory Director, Program Office USAID Tablas, Víctor Program Analyst UNDP 57 Name Position Organization Ventura, Alan Jesús Data Quality Specialist, Quality Control Team MEL Initiative Vides, Carolina Backstop, Democracy and Governance, Program Office USAID Martinez, Yosselin Beatriz Database Development and Processing Specialist, Quality Control Team MEL Initiative Zogbi, Elisa Deputy Director, Office of Democracy and Governance USAID El Salvador/Region Honduras Andrews, Héctor Regional Program Coordinator, Program Office USAID Castellanos, Julieta Director UTECI UNAH Castillo, Carlos Andrés Project Management Specialist, Democracy and Governance Office USAID Chaker, Alissar Deputy Resident Representative UNDP David, Alexander Sub-Director INE Hernández, Carlos Executive Director ASJ Junibeth Pinto, Blanca Technician UGAGLO Lanza, Carmela InfoSegura Project Coordinator UNDP Mejía, Alejandra Municipal Liaison Technician Observatory of Coexistence and Public Safety of San Pedro Sula Muñoz, Marlene Coordinator Observatory of Coexistence and Public Safety of San Pedro Sula Ponce, Graciela Executive Director INE Ramírez, Lester Director of Governance ASJ Rodero, Ernesto Coordination Specialist, Office of the Resident Coordination United Nations Systems Sagastume, Donald Technical Coordinator Sub-Secretariat of Inter￾institutional Affairs, Secretariat of Security Servellón, Elsa Director CENISS Sierra, Marcio Enrique Director CEASCI Sosa, Ángela Assistant to the Executive Director CEASCI Guatemala Amado, Anabella Investigator Institute of Investigation and Projection on the State, Rafael Landívar University Arias, Rebecca Resident Coordinator / Resident Representative UNDP Assivillega, Brenda Edith Coordinator of Observatory UPCV, Governorate of Chimaltenango Azurduy, Luis Technical Team Leader for Citizen Security, Office of Democracy and Governance USAID 58 Name Position Organization Barrios, Karin Coordinator INE Berteton, Mynor Head of Statistics, General Directorate PNC Díaz, Ana María Office Director UNDP Díaz Molina, Eduardo Crime and Violence Analyst UNDP Gaspar de León, Darío Chief, Joint Operations Division, General Directorate PNC López, Claudia V. Official of Population and Development UNFPA Mendoza, Silvia Statistical Technician UNDP Milian, Spencer InfoSegura Activity Manager USAID Puac, Dougly Elena Observatory Analyst UPCV, Governorate of Chimaltenango Romero, Axel Third Vice Minister of Government and Prevention of Violence and Crime MINGOB Rosales, Carlos Citizen Security, Office of Democracy and Governance USAID Sanabria, Claudia Program Officer, Office of Governance UNDP Villatoro, Mariela A. Technical Adviser of the Third Vice￾Ministerial Office of the Interior MINGOB Azurduy, Luis Leader Technical Team for Citizen Security, Office of Democracy and Governance USAID/Guatemala López, José USAID Regional Coordinator in Guatemala USAID/Guatemala BIBLIOGRAPHY OF DOCUMENTS REVIEWED Carcach, Carlos A. 2016. “Crime and Punishment in El Salvador.” The Encyclopedia of Crime and Punishment. John Wiley and Sons. CONOSE. 2018. “Technical Protocol for Field Research in Violent Contexts.” Consejo Nacional de Seguridad Ciudadana. 2018. Informe de Monitoreo, 2016-2017: Plan El Salvador Seguro. Correa-Cabrera, Guadalupe. 2017. “Crimen, corrupción y pérdida del monopolio de la violencia en Honduras: Reflexiones y apuntes de viaje a San Pedro Sula, Choloma, El Progreso y la Ceiba.” Wilson Center. The Economist. 2018. “The Northern Triangle Is Becoming Less Murderous.” December 8. Eguizábal, Cristina, Matthew C. Ingram, Karise M. Curtis, Aaron Korthuis, Eric L. Olson, Nicholas Phillips. 2015. Crime and Violence in Central America’s Northern Triangle. Wilson Center. Fundación Arias para la Paz y el Progreso Humano n.d. Sistematización Regional Externa de la Experiencia de Gestión de la Información InfoSegura. n.d. Línea de Base: Belice. n.d. Línea de Base: Costa Rica. n.d. Línea de Base: El Salvador. 59 n.d. Línea de Base: Guatemala. n.d. Línea de Base: Honduras. n.d. Línea de Base: Panamá. n.d. Línea de Base: República Dominicana. Gobierno de la República de Guatemala 2018 Compendio Estadístico 2018: “Indicadores de Convivencia y Seguridad Ciudadana.” 2018 Herramientas para la priorización y focalización de políticas en materia de seguridad ciudadana: Índice de Priorización Municipal, Guatemala. Gobierno de la República de Honduras 2016 Campaña de Comunicación. Secretaría de Seguridad a través de la Sub Secretaría de Asuntos Interinstitucionales de la Unidad Técnica de Coordinación Interinstitucional (UTECI). n.d. Del Dato a la Acción: Todas las historias cuentan. Secretaría de Seguridad a través de la Sub Secretaría de Asuntos Interinstitucionales de la Unidad Técnica de Coordinación Interinstitucional (UTECI). n.d. Informe Dato Oficial de Homicidios, Suicidios y Muertes por Lesiones de Tránsito en Honduras. 1 de enero al 31 de diciembre de 2016. Sub Secretaría de Asuntos Interinstitucionales, Unidad Técnica de Coordinación Interinstitucional (UTECI) n.d. Informe Dato Oficial de Homicidios, Suicidios y Muertes por Lesiones de Tránsito en Honduras. I semestre 2016. Sub Secretaría de Asuntos Interinstitucionales, Unidad Técnica de Coordinación Interinstitucional (UTECI) n.d. Informe Dato Oficial de Homicidios, Suicidios y Muertes por Lesiones de Tránsito en Honduras. I semestre 2017. Sub Secretaría de Asuntos Interinstitucionales, Unidad Técnica de Coordinación Interinstitucional (UTECI) n.d. Informe Dato Oficial de Homicidios, Suicidios y Muertes por Lesiones de Tránsito en Honduras. I semestre 2017. Sub Secretaría de Asuntos Interinstitucionales, Unidad Técnica de Coordinación Interinstitucional (UTECI) n.d. Informe Dato Oficial de Homicidios, Suicidios y Muertes por Lesiones de Tránsito en Honduras. I enero – 31 de marzo de 2017. Sub Secretaría de Asuntos Interinstitucionales, Unidad Técnica de Coordinación Interinstitucional (UTECI) n.d. Informe Dato Oficial de Homicidios, Suicidios y Muertes por Lesiones de Tránsito en Honduras. Enero – diciembre de 2017. Sub Secretaría de Asuntos Interinstitucionales, Unidad Técnica de Coordinación Interinstitucional (UTECI) n.d. UTECI: Manual de Comunicación Interna. 2017 Boletín Mesa de Protección Integral de la Niñez. Ministerio Público. Municipalidad de San Pedro Sula. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Choloma. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Choluteca. Boletín No. 1, enero-junio 2016-2017. 60 n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Progreso. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Morazán. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Puerto Cortéz. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Santa Bárbara. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de SantaCruz de Yojoa. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana deTrujillo. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Villa Nueva. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de San Pedro Sula. Boletín No. 1, enero-junio 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Catacamas. Boletín No. 2, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Choloma. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Comayagua. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Santa Bárbara. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana del Distrito Central. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de El Progreso. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Gracias. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Juticalpa. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de La Ceiba. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de La Lima. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de La Paz. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Macuelizo. Boletín No. 1, enero-diciembre 2016-2017. 61 n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Morazán. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Nueva Arcadia. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Olanchito. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Puerto Cortés. Boletín No. 2, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Quimistán. Boletín No. 2, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de San Manuel Cortés. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de San Pedro Sula. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Santa Bárbara. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Santa Cruz de Yojoa. Boletín No. 2, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Santa Rosa de Copán. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Siguatepeque. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Tocoa. Boletín No. 1, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Trujillo. Boletín No. 2, enero-diciembre 2016-2017. n.d. Boletín Informativo del Observatorio Municipal de Convivencia y Seguridad Ciudadana de Villa Nueva. Boletín No. 1, enero-diciembre 2016-2017. 2018 Violencia Homicida en el Contexto Social de la Convivecia. Boletín CEASI No. 1. International Crisis Group 2014 Corridor of Violence: the Guatemala-Honduras Border. Latin America Report No. 52. 2017 Mafia of the Poor: Gang Violence and Extortion in Central America. Latin America Report No. 62. 2017 El Salvador’s Politics of Perpetual Violence. Latin America Report No. 64. 2018 Saving Guatemala’s Fight Against Crime and Impunity. Latin America Report No. 70. n.d. Life Under Gang Rule, El Salvador. InSight Crime. 2017. “Guatemala Profile.” Nov. 29. Meyer, Peter J. 2015. “U.S. Strategy for Engagement in Central America”: Policy Issues for Congress. Congressional Research Service. 62 Programa de Naciones Unidas para el Desarrollo (PNUD)/United Nations Development Programme (UNDP) n.d. Ejercicio de Priorización Municipal para la Toma de Decisioes en Política Pública con Enfoque Mutidimensional en Honduras. n.d. M&E Indicators. n.d. M&E Progress 2017. n.d. Performance Indicator Reference Sheet, 2014-2017. n.d. Performance Indicator Reference Sheet, 2017-2020. n.d. Theory of Change 2014-2017. n.d. Theory of Change 2017-2020. 2014 Annual Implementation Work Plan. August. 2015 Annual Implementation Work Plan, Year II. November. 2016 Annual Implementation Work Plan, Year III. 2017 Substantive Revision. May. 2018 Annual Report 2017-2018. InfoSegura Project. October. 2018 Avance de implementación del proyecto InfoSegura, Oct. 10. 2018 Annual Work Plan, January-December, 2018 n.d. Project Document (InfoSegura) n.d. Project Proposal: “Evidence-Based Information Management for Citizen Security in Central America.” InfoSegura. United Nations Office on Drugs and Crime. 2016. World Drug Report. United States Agency for International Development (USAID) n.d. U.S. Strategy for Engagement in Central America. 2014 Plan of the Alliance for Prosperity in the Northern Tringle: A Road Map. n.d. Regional Development Cooperation Strategy 2015-2019. Central America and Mexico (CAM). 2016 CAM Regional Develoopment Cooperation Strategy (RDCS). 2017 ADS Chapter 205: Integrating Gender Equality and Female Empowerment in USAID’s Program Cycle. 2017 Annual Portfolio Review. June 11. 2018 “Compendio: Investigaciones basadas en evidencia para Políticas Públicas en Seguridad Ciudadana en Guatemala.” 2018 Monitoring and Evaluation System Assessment of USAID Regional Security Activity: InfoSegura. 2018 ADS Chapter 201: Program Cycle Operational Policy. United States Agency for International Development (USAID) and United Nations Development Programme (UNDP) 2014 Minuta de Reunión Comité de Proyecto InfoSegura. 14 de octubre. 2015 Informe Anual InfoSegura. Octubre 2014-Septiembre 2015. n.d. Informe Intermedio de Actividades. Octubre 2016 – Mayo 2017. 63 n.d. Informe Semestral, Unidad de Coordinación. Octubre 2015 – Marzo 2016. 2015 Reporte annual, Unidad de Coordinación. Septiember 2014 – Marzo 2015 n.d. Informe Annual InfoSegura. Octubre 2015 – Septiembre 2016 2016 “El Valor del Dato Consensuado.” Estrategia de Comunicación. Unidad Técnica de Coordinación Interinstitucional (UTECI). Honduras. 2016 Los Observatorios Municipales de Covivencia y Seguridad Ciudadana de Honduras: De la Recolección al Uso de Información en Políticas Púbicas. 2016 Minuta de Reunión Comité de Proyecto InfoSegura. 24 de junio. 2016 Protocolo para la Reconciliación de Datos: Mesas temáticas: Criminalidad, Muertes Violentas, Convivencia, Violencia Intrafamiliar, Doméstica, Delitos Sexuales￾Maltrato, Victimización, Prevención de Violencia. Honduras. 2017 Annual Report 2016-2017. November 2017. 2017 Bi-Annual Report. October 2017 – March 2018. May. 2017 El Continuum de la violencia contra las mujeres en la región centroamericana. 2017 Factores intervinientes para la reproducción y el rompimiento del círculo de la violencia en la vida de las mujeres. Presentación de resultados de investigación. 2017 Informe de Revisión de Medio Término del Proyeto InfoSegura. Agosto￾Noviembre. Consultor Externo, Nana Gibradze 2017 Minuta de Reunión Comité de Proyecto InfoSegura. 21 de agosto. 2018 Informe Anual InfoSegura. 2017-2018. Octubre. n.d. Informe Intermedio de Actividades: Gobernbilidad y Construcción de Paz—Centro Regional—LAC. Octubre 2016-Mayo 2017. n.d. Índice Alianza por la Paz y la Justicia (APJ). Poder Judicial, Línea de Base. Honduras. n.d. Monitoring and Evaluation Plan: Evidence-Based Information Management for Citizen Security in Central America, InfoSegura. Watkins, Ali, Meredith Kohut. 2018. “A Conflicted War: MS-13, Trump and America’s Stake in El Salvador’s Security.” The New York Times. December 10. World Bank. 2009. Guatemala Poverty Assessment. Report No. 43920-GT 64 ANNEX VII: STATEMENT OF DIFFERENCES – INFOSEGURA PROJECT STATEMENT OF DIFFERENCES INFOSEGURA PROJECT General observations on the InfoSegura Regional Project Performance Evaluation Multiple findings, conclusions and recommendations in the InfoSegura Regional Project Performance Evaluation conducted by the Evaluation Team note an absence of a criminological approach, civil servants at different institutions lack capacity for criminal analysis, and the absence of a theoretical framework in analyzing the issue of insecurity. In this regard, it is worth recalling that the InfoSegura Regional Project was designed and is being implemented on the basis of the United Nations paradigm for citizen security, which places people at the center of a multidimensional approach. Understood to be a public asset, citizen security refers to a democratic citizen order that eliminates the threat of violence in the population and allows for safe and peaceful coexistence. It involves the effective enforcement of a portion of the spectrum of human rights, specifically the right to life, to personal safety, gender equity and other rights inherent to the personal sphere (inviolability of the home, freedom of transit, etc.) as well as the enjoyment of assets. This approach to citizen security expands on conceptions that define security purely in terms of crime and offenses. From this multidimensional approach, the Project supports the development of national capacities for information management as input for the design, support for implementation, monitoring and evaluation of public policies on citizen security that are gender-sensitive, from a territorial perspective, emphasizing the social and situational prevention of the crime, its control and criminal prosecution, victim care and protection, and the rehabilitation and reinsertion of people that have violated criminal law. Specific comments: EQ1: To what extent has InfoSegura developed the capacity within STUs (located within each country’s respective Security Ministry) to collect, record, and use crime and violence data for analysis? EL SALVADOR There is a need to revise the methodology and conduct of the Crime Victimization Survey to ensure that it is based on a properly constructed sampling frame to avoid coverage error. (page iii) The report notes that the survey-data collection methodology needs revision. These methodological recommendations have been taken into account in the 2019 Survey, which includes the updated mapping. The survey report, moreover, can be improved to ensure it contains relevant findings and sufficient information for users to assess the data quality. Other improvements can be made to the structure and writing of the report. (page iii) 65 As regards findings and conclusions on the quality of the Victimization Survey and particularly reporting, this is in line with the role and mandate of the DIGESTYC, namely, information gathering, and not analysis. Nevertheless, the corresponding analyses are being processed by the CONOSE Network, in coordination with the MJSP. To date, all efforts have been devoted to homicide counts. At the same time, inconsistencies exist in homicide data reported by the PNC crime analyzer and the DIA website and by the crime analyzer and the PNC’s Unit of Access to Public Information (UAIP). (page iii) Since 2019, the board for harmonization has also included the crime of femicide (reference in Spanish): http://www.seguridad.gob.sv/dia/estadisticas-homologadas/repositorio-de-los￾homicidios-desagregados-por-las-variables-homologadas/ Since the DIGESTYC survey is the first source of official data on the topic, the unit record data should be made available to the public to allow its use by researchers and public policy analysts. The same applies to the data files of the Violence Against Women Survey (page ix and 17). The 2017 and 2018 Culture of Peace Surveys are already on line at the DIA-MJSP website (Spanish only). http://www.seguridad.gob.sv/dia/monitoreo-y-evaluacion/encuesta-de-victimizacion/ The Survey on Violence against Women is on line at: http://aplicaciones.digestyc.gob.sv/observatorio.genero/eviolencia2018/index.aspx EQ2.1: What are the internal and external factors that foster or impede the governments to use data? HONDURAS Some elements that impede the use of data for policymaking are linked to the lack of trained professionals in fields related with the sociology of crime and criminology (page v) The Instituto Universitario en Democracia, Paz y Seguridad (IUDPAS) and the Secretaría de Seguridad, trained 120 public employees in the country's security and justice sector by way of a university certification program in Criminalistics and Criminology, with a forensic medicine focus. This certification program strengthened technical skills in prosecutors, forensic medical personnel, Dirección Policial de Investigaciones (Police Investigation Directorship) and the Subsecretaría de Seguridad en Asuntos Interinstitucionales de la Secretaría de Seguridad (the Secretary for Security's Sub-Secretariat for Inter-Institutional Affairs), providing them with knowledge that enables them to identify new criminalistic-forensic techniques and crime scene processing techniques to adequately handle evidence, making use of disciplines such as criminalistics, documentation, forensic ballistics, marks and patterns, toxicology, serology, anthropology, forensic psychiatry, and others. 66 4.4.1 EQ2: To what extent have Central American governments used (as planned) crime and violence data—produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions? GUATEMALA Government staff in different institutions, including the PNC, noted that the insufficient capacity for analysis in different state entities which could be making use of available data impedes the more effective use of data for policymaking in El Salvador (Page 42) Should say: policymaking in Guatemala EL SALVADOR, GUATEMALA AND HONDURAS EQ3.1: What are the internal and external factors that foster or impede collaboration among the members of CONOSE? The absence of research within the framework of criminology and the absence of training opportunities in criminological approach. The CONOSE member-institutions did not include criminological analysis as a priority in the research and training agenda they proposed. 4.1.1 EQ1: To what extent has InfoSegura contributed to changing the level of quality and timeliness of crime and violence data definition, collection, and recording in the region? GUATEMALA The survey was funded by the Ministerio de Gobernación (Governance Ministry, or MINGOB) and designed by the Instituto Nacional de Estadísticas (National Institute of Statistics or INE), while data was collected by a private company and processed with the assistance of the Mexican Instituto Nacional de Estadística, Geografía e Informática (National Institute of Statistics, Geography, and Informatics, or INEGI). InfoSegura supported the survey through the United Nations Office on Drugs and Crimes’ (UNODC) Center of Excellence and with the publication of the findings. At the time of this writing, the survey findings have not been published, pending MINGOB approval (Page 11) InfoSegura and the Centro de Excelencia (INEGI Center for Excellence) jointly provided technical assistance in preparing conditions specifically in survey design, enumerator training, accompaniment for data collection, and the final phase of data analysis, which was performed under the framework agreement with UNODC. Currently, deep analysis is being performed in 67 the context of the Mesa Tecnica Institucional (Inter-institutional Technical Roundtable) and results are expected to be shared over the course of this government administration. 4.4.1 EQ2: To what extent have Central American governments used (as planned) crime and violence data—produced by InfoSegura and by other institutions that have received technical assistance under InfoSegura—as input to make effective citizen security-related policy decisions? At the time of writing, the survey findings have not been published pending MINGOB approval (Página 11) The principal challenge at this point is the publication of the ENPEVI, whose delay is attributable to the tense and volatile political situation in Guatemala, intensified by the upcoming electoral period in 2019 (Page 22). The Ministry of Governance is analyzing ENPEVI results in the context of the Mesa Tecnica Interinstitucional (Inter-Institutional Technical Roundtable), and data is being formalized and certified by INE. Results are expected to be shared over the course of this government administration. Annex II: El Salvador Profile The latest effort to address the problem of violence is a project called Plan El Salvador Seguro (“Safe El Salvador Plan”) under the Sánchez Cerén government. Although the security policy has continued to focus on law enforcement, the project has initiated some violence-prevention measures. The project, costing about (US$) 200 million per year and financed mostly with donor funds and to a lesser extent with a special Salvadoran tax, began by working in municipalities most affected by violence. (page 46. Annex 2).65 Plan El Salvador Seguro is not a project, rather it is public policy. It is a road map providing a framework for the efforts of the institutions that are responsible for citizen security in the country. There are 5 Axis: Violence Prevention, Crime Control and Prosecution, Rehabilitation & Social Reinsertion, Victim Assistance & Protection, and Institutional Strengthening. Included are all actions at the national and local levels. At the local level, work started in 50 municipalities identified in a municipal prioritization index prepared with InfoSegura support. State institutions are responsible for implementation of the Plan, with other social stakeholders participating. At the time of formulating the PESS, it was estimated that over a 2-billion-dollar investment was required for implementation. To the date, it has been impossible to calculate the precise figure of the annual investment, since it includes funding from the budgets of the public institutions, from international donors, and from the special tax. According to some calculations, the lion's share of investment in violence prevention has come from the international donors. In November 2015, the Legislative Assembly approved the “Ley de Contribución Especial para la Seguridad Ciudadana y Convivencia” (Law for Special Contributions for Citizen Security and Coexistence or CESC) and the “Ley de Contribución Especial a los Grandes Contribuyentes” (Law for Special Contribution from the Large Taxpayers), in the interest of funding actions for citizen 65 Of the $93 million collected in taxes in 2017, around 70 percent went to finance the police and the armed forces. 68 security and coexistence. In late 2018, a total of $297,155,897.00 was levied, 24.7% of which was invested in violence prevention, 43.1% in actions related to crime control and criminal prosecution; 6.1% in the Axis for rehabilitation and social reinsertion; 2% in victim assistance, and 24.1% went to institutional strengthening. The National Council on Citizen Security and Coexistence issued recommendations on the distribution of the funds. It has suggested that 73% of the investment go to prevention, as established in the Plan El Salvador Seguro. It also deems a financial evaluation is necessary to determine whether the economic incentives that law enforcement agents currently receive from CESC, like bonuses, can become part of their wages and/or propose other initiatives for work benefits to become part of the General Budget of the Nation. 69 U.S. Agency for International Development 1300 Pennsylvania Avenue, NW Washington, DC 20523