Performance Evaluation of USAID/Mexico’s Crime and 
Violence Prevention Activity:
Juntos para la Prevención de la Violencia
August 2020
This publication was produced at the request of the United States Agency for International Development. It was
prepared independently by the Center on Conflict and Development, Texas A&M University. 
Photo credit: JPV 
CONTENTS
ACRONYMS ...............................................................................................................................................................................3
EXECUTIVE	SUMMARY..........................................................................................................................................................5
I. EVALUATION PURPOSE ...........................................................................................................................................10
II. PROGRAM	BACKGROUND .......................................................................................................................................11
III. EVALUATION	METHODS	AND	LIMITATIONS ....................................................................................................13
III.1 STAGE 1 METHODOLOGY................................................................................................................................................................... 13
III.2 STAGE	2 METHODOLOGY................................................................................................................................................................... 16
STAGE 1: WERE LPSS	STRENGTHENED AND EVIDENCE-BASED INFORMED INTERVENTIONS
IMPLEMENTED?..................................................................................................................................................................................... 19
IV.1 JPV THEORY	OF	CHANGE: THEORY	AND	PRACTICE........................................................................................................................ 20
IV.1.1 JPV expenses...................................................................................................................................................................................25
IV.2 JPV CONTRIBUTION TO	INSTITUTIONAL	CAPACITY-BUILDING .................................................................................................... 27
IV.3 JPV CONTRIBUTION	TO	THE	GENERATION	OF	MULTISECTOR	PREVENTION	AGENDAS............................................................. 31
IV.3.1 Multisector groups and government participation ................................................................................................................32
IV.3.2 Anchor and strategic actors.........................................................................................................................................................35
IV.3.4 Private sector...................................................................................................................................................................................37
IV.3.5 Civil society.......................................................................................................................................................................................38
IV.3.6 Consultants, think tanks, and academic institutions.............................................................................................................39
IV.3.7 JPV organizational structure........................................................................................................................................................40
IV.3.8 USAID...............................................................................................................................................................................................41
IV.4 JPV CONTRIBUTION	TO	THE	IMPLEMENTATION	OF	EVIDENCE-BASED	PREVENTION	INTERVENTIONS ................................. 42
IV.4.1 Key avenues for reform: what the evidence says..................................................................................................................43
IV.4.2 JPV Donations program................................................................................................................................................................44
IV.4.3 Programmatic quality of JPV-supported interventions...........................................................................................................45
Key Principle 1: Concentration................................................................................................................................................................45
Key Principle 2: Coordination ..................................................................................................................................................................54
Key Principle 3: Proactivity.......................................................................................................................................................................54
IV.5 JPV’S	CONTRIBUTION	TO	THE	SCALE-UP	OF	EFFECTIVE	MODELS ............................................................................................... 56
IV.5.1 Pathways to scale and elements of effectiveness..................................................................................................................57
IV.5.2	Scaling-up	challenges..............................................................................................................................................................60
STAGE 2—DID THE PROGRAM CONTRIBUTE TO A REDUCTIONOF AT-RISK YOUTHLIKELIHOOD
PARTICIPATING INCRIME AND VIOLENCE? ............................................................................................................................. 68
IV.6 COVERAGE	AND	TARGETING ............................................................................................................................................................. 69
IV.7 JPV-SUPPORTED	INTERVENTIONS	RESULTS	IN	CRIME	AND	VIOLENCE	REDUCTION	IN	TARGETED	LOCALITIES ................... 72
V.	RECOMMENDATIONS .....................................................................................................................................................75
REFERENCES ..........................................................................................................................................................................79
ANNEX	I. PROGRAM	BACKGROUND........................................................................................................................................................... 81
ANNEX	II. DATA	COLLECTION	INSTRUMENTS.......................................................................................................................................... 82
ANNEX	III. RESULTS	ACHIEVED	BY	MEASSURMENT	INSTRUMENT ....................................................................................................104
ANNEX	IV. COMPARATIVE	MUNICIPALITIES	MATRIX...........................................................................................................................108
ANNEX	V. DONATIONS	PROGRAM...........................................................................................................................................................117
ANNEX	VI. JPV ORGANIZED	EVENTS	AND	STUDY	TOURS ....................................................................................................................121
ANNEX VII. JPV’S	RESOURCES	DEVOTED	TO	TRANSFER	AND	SCALE-UP	AND	RESULTS..................................................................128
ANNEX	VIII. TARGETING	ANALYSIS .......................................................................................................................................................134
ANNEX	IX. PRELIMINARY	STATISTICAL	ANALYSIS ...............................................................................................................................138
3
ACRONYMS
5Rs resources, roles, relationships, rules, results
BAM Becoming a Man
CAS Center of Social Services (Spanish acronym)
CBT cognitive behavioral therapy
CCSJ Citizen Council for Security and Justice (Spanish acronym)
CEDAT Centro de Desarrollo y Atención Terapéutica 
CIJ Center for Youth Integration (Spanish acronym)
CINLAAC Civic Council of the Institutions of Nuevo Leon (Spanish acronym)
CJHM Civic Justice Homologated Model
CLA collaborating, learning, and adapting
COLEF College of the Northern Border (Spanish acronym)
COPARMEX Confederación Patronal de la República Mexicana (Commission of Large Enterprises 
of the Mexican Republic)
CONAPO National Population Council (Spanish acronym)
ConDev Center on Conflict and Development
COPREV Council for Crime and Violence Social Prevention
CPTED crime prevention through environmental design
CSO civil society organization
DIF National System for the Family Integral Development (Spanish acronym)
DOI Development Objective I
EBP Best-Practices Scale
ENSU National Survey of Urban Security (Spanish acronym)
FECHAC Fundación del Empresariado Chihuahuense
FICOSEC Fideicomiso para la Competitividad y la Seguridad Ciudadana
FORTASEG Strengthening Security Program (Spanish acronym)
GBV gender-based violence
GOPI-P Government Performance Index—Prevention
IDIQ indefinite delivery–indefinite quantity
INEA National Institute for Adult Education (Spanish acronym)
INEGI National Institute for Statistics and Geography (Spanish acronym)
IRB Institutional Review Board
IRLSP Local Prevention System Performance Index (Spanish acronym)
J-PAL Abdul Latif Jameel Poverty Action Lab 
JPV Juntos para la Prevención de la Violencia (Together to Prevent Violence)
LPS local prevention system
M&E monitoring and evaluation
NGO nongovernmental organization
OPI-P Organization Performance Index—Prevention
PoP problem-oriented policing
RFP request for proposal
ROLE Resources, Options, and Freedom of Election (Spanish acronym) 
SAVRY Structured Assessment of Violence Risk in Youth
SIMOPIN Offender Monitoring System (Spanish acronym)
SIPINNA National System for the Protection of Girls, Boys, and Adolescents (Spanish acronym)
SESNSP Secretary General of National Public Security (Spanish acronym)
STYL Sustainable Transformation of Youth in Liberia
TCO transnational criminal organization
4
USAID United States Agency for International Development 
UVAQ Universidad Vasco de Quiroga
YTT Youth-Targeting Tool
5
EXECUTIVE SUMMARY
The purpose of this performance evaluation is to gauge the effectiveness of the United States 
Agency for International Development (USAID) program Juntos para la Prevención de la Violencia
(JPV), or “Together to Prevent Violence,” in promoting reductions in levels of crime and violence
and gather recommendations for improved future programming. In particular, this evaluation 
assesses the effectiveness of JPV in achieving the intermediate results of strengthening local 
prevention systems (LPSs) and in reducing the likelihood of youth participating in crime and 
violence. It also explores JPV’s contribution to the development objective of sustainably 
decreasing crime and violence in the targeted communities. 
To answer the primary questions of the evaluation, we pursue a two-stage approach. In Stage 1,
we analyze the results achieved in strengthening local systems. We identify implemented 
prevention interventions, as well as the local actors and beneficiaries who were supported by 
JPV. In Stage 2, we study whether the interventions contributed to: a) a reduction in the 
likelihood of youth engaging in crime and violence, and/or b) an overall reduction in crime and 
violence in targeted municipalities. The results presented for Stage 2 are preliminary, because the 
focus group interviews and youth survey have been delayed due to COVID-19 and will be 
completed upon a safer environment adhering to USAID’s “No Harm Done Principle.” The 
categorized preliminary findings to respective questions are:
1. To what extent was the design of the JPV program relevant and sufficient for 
achieving the stated results and contributing to reduced crime and violence in 
the targeted communities?
a. To what extent did JPV’s updated theory of change (i.e., targeting high-risk 
youth and local systems approach) hold true?
b. What JPV program elements were the most effective and least effective at 
promoting reduction in crime and violence in targeted communities?
JPV’s theory of change was successfully reflected in the implemented strategy, which focused on 
working with local actors to fulfill three main functions of local systems: 1) create knowledge, 2) 
implement solutions, and 3) improve practices. This local system-based framework evolved during 
the life of the program. The implementation commenced from the first function and gradually 
moving to the third. Because the practical implementation of JPV was launched early 2017, it did 
not have enough time to generate stronger improved practice outcomes. To summarize, the 
program achieved stronger results in developing a shared understanding of the local violence 
phenomenon and in developing general organizational capacities at the local level, as well as 
moderate results in identifying promising practices and in implementing evidence-based models. 
Due to lack of evaluation of implemented solutions that are necessary to develop appropriate 
lessons and impact policy, improved practices were not as apparent. 
JPV implemented a transversal institutional-strengthening strategy and four substantive 
transversal strategies: a) solutions for at-risk youth, b) civic justice, c) gender strategy, and d) 
environmental design prevention strategy. The first two strategies achieved stronger results for 
6
the third function – improved practices. The following questions explore in more detail which 
components worked better and why.
2. What evidence is there that JPV programming has strengthened LPSs in a 
sustainable manner?
a. To what extent are the local actors capable of maintaining the improved 
crime prevention policies, to the extent that there are improved policies), 
without USAID support?
b. Have key avenues for reform and a reduction in crime and violence in the 
targeted communities been addressed through JPV programming and 
implementation? Why or why not?
JPV achieved positive results with the implementation of the transversal institutional￾strengthening strategy. The targeting of activities was too broad and involved a lot of local 
implementing partners. This policy had an opportunity cost of gaining stronger results with 
possibly fewer organizations implementing more focused programs. Even though the program 
had notable success in increasing awareness and positioning the crime-and-violence-prevention 
agenda within the local and national systems; the results achieved in terms of generating 
sustainable, multisector, coordinated agendas were mixed. JPV succeeded at enabling multisector 
collaboration when prevention strategies aligned with interest groups, and government 
participation was obtained when related short-term milestones were specified and followed. 
JPV opted to create “anchor actors” and “strategic actors,” with the objective of ensuring 
sustainability by empowering these actors to become articulators of collaborative efforts and to 
help other members develop their capacities. We find that this strategy was accompanied by
several challenges that limited its success, particularly because the program failed to create the 
conditions for these actors to become the leaders of local agendas in most cases. Moreover, our 
analysis shows that the conceptualization of the anchor actor, while attractive, lacked appropriate 
coordination and application. This role did not exist or was not necessary in all systems. Finally, 
we find that while JPV’s organizational structure with regional coordinators was very successful 
in increasing local actors’ participation, the hierarchical structure of USAID may have hindered 
local ownership. In sum, we find that the prevention agenda cannot always be clustered within a 
single group and that specialization and working with diverse leadership are desirable.
In terms of the implementation of evidence-based policies, we assess whether JPV supported 
interventions that followed the three principles of effective violence prevention found in the 
literature: concentration, coordination, and proactivity. We find that JPV technical assistance 
activities added to the concentration principle by supporting the development of civil society
organization (CSO) programs with enhanced design, delivery, and coverage, although pressing 
challenges remain in terms of addressing youth risks and needs to prevent recidivism. JPV also 
contributed, but to a lesser extent, to the coordination principle by enhancing communication 
between CSOs and other relevant stakeholders, although major challenges prevailed with 
coordination with public institutions. The proactivity principle remains an area of opportunity for 
future USAID efforts. 
7
JPV evolved from a broader strategy looking at identifying promising interventions to supporting 
reintegration strategies to, finally, focusing almost exclusively on strengthening and transferring 
cognitive behavioral therapy (CBT) models and the Civic Justice Homologated Model (CJHM).
Through its evolution, JPV made good progress in identifying promising practices, but achieved 
less in strengthening sustainable models. Therefore, the program suffers from generating evidence 
of effectiveness for replication in most cases. JPV was successful at leveraging different pathways 
to scale to expand two CBT models, Resources, Options, and Freedom of Election (ROLE being 
its acronym in Spanish), Mente Libre, and the CJHM. We identify that the focus from the beginning 
on scale, the provision of a meticulous strengthening process through access to overarching 
resources, and the generation of relationships of trust and co-investments with partners are key 
to achieving scale. Although JPV was very successful in supporting the adoption of these models, 
rushed expansion, absence of rigorous evaluations, weak technical capabilities in implementing 
organizations, and lack of consideration for differences in local circumstances posed challenges 
for model fidelity and effectiveness.
3. What evidence is there that JPV has reduced the likelihood of at-risk youth in 
targeted communities participating in crime and violence?
a. What evidence is there that at-risk youth reached with JPV programming 
will engage in alternative, positive pathways and steer away from 
involvement in crime and violence, including with transnational criminal 
organizations (TCOs)?
b. Is JPV’s coverage or reach of at-risk youth in targeted communities (i.e., 
number of at-risk youth reached) sufficient to initiate a detectable 
reduction in youth involvement in crime and violence? Why or why not? 
c. To what extent have gender-based roles, including but not exclusive to the 
mothers of at-risk youth, affected the ability of the interventions to achieve 
their stated results (effectiveness)?
Our field research and evidence review suggest that crime in Mexico is a highly contextual, socio￾culturally driven behavioral phenomenon. Therefore, a successful intervention also needs to take 
into account the broader context of violence to which program participants continue to be 
exposed during and after the program. However, our amassed secondary information shows that
JPV-supported interventions usually only targeted 1 out of the 10 most violent neighborhoods
within the municipalities of interest, and often targeting only a limited number of youths in these 
neighborhoods. Our preliminary results suggest that, while the intervention may have had a 
positive effect on individual participants in those municipalities, a wider coverage, both in terms 
of services offered and youth reached, would be necessary in order to obtain community-level 
effects in crime and violence reduction.
In terms of services offered, we find that the designs of existing interventions do not sufficiently 
tackle youth risk factors and that gender roles are barely addressed. The programmatic offer
currently covers impulsive behavior, anger management, and employability skills, and to a lesser 
extent substance consumption. However, data from the Youth-Targeting Tool (YTT) confirm 
that a wider range of interventions is needed to address youth risk factors, particularly to 
promote protective factors. Interventions that are able to promote protective factors and 
8
interrupt the community and family dynamics that produce antisocial behaviors remain urgent, as 
does expanding the offer of substance abuse treatment.
These results are preliminary. At a later stage, we will perform focus groups with the beneficiaries 
of direct interventions to understand which components might have worked or not and why. We 
will also collect data of at-risk youth in the targeted municipalities through a representative 
sample of communities with high homicide rates to measure individual-level exposure and 
attitudes toward violence. 
4. What evidence is there that the USAID/Mexico crime-and-violence-prevention 
programming through the JPV activity has contributed to achieving the goal of 
reducing crime and violence within the targeted JPV communities?
a. Where and what are examples of the JPV activities contributing to reduced 
levels of crime and violence in the targeted municipalities?
b. To what extent have citizen perceptions of security increased in targeted 
municipalities?
We perform a descriptive statistical analysis of existing secondary data on homicide and other 
crime rates, as well as on citizen perceptions of security, to investigate JPV’s role in reducing 
crime and improving perceptions. However, we remain very cautious and refrain from drawing 
any causal connections from this analysis. The increased levels of violence in Mexico are in large 
part attributable to organized crime. Consequently, insecurity perceptions are also driven by 
violence related to drug cartels and to other highly visible acts of violence that affect perceptions. 
Hence, a decrease in homicide rates and reducing insecurity perceptions, may require different 
strategies than the one envisioned for JPV and that entail longer-term institutional reform and 
more traditional law enforcement action.
Our initial statistical analysis shows that homicide rates present positive trends for all 
municipalities of interest since 2015. Insecurity perceptions have risen in all municipalities, except 
for Guadalupe and Escobedo (although data for these municipalities are only available since June 
2018). We also explore other available indicators related to antisocial behaviors that may be 
better suited to explore the program’s contribution to reducing the likelihood of youth engaging 
in crime and violence. Unfortunately, the lack of clear counterfactual information, along with 
limited socio-economic data, hindered our ability to isolate the effect of the program on youth 
population. From our field explorations we generate a plausible hypothesis that JPV could have 
impacted individual youth perceptions, attitudes, and exposure to violence. In order to address 
the challenges with the existing data, we have designed a unique and tailored survey of youth that 
will measure youth exposure, attitudes, and behaviors toward violence. Our stratified sampling
procedure will include youth in the targeted municipalities as well as youth from similar 
municipalities which remained outside the realm of JPV, to form a comparison group.
Based on JPV’s learnings, we provide guidelines for more impactful, proven, and sustainable
approaches for future crime-and-violence-prevention programs. First, crime-and-violence￾prevention systems should be narrowly defined according to expected outcomes rather than 
long-term term development goals. This is particularly relevant for issues as complex as crime 
and violence and in settings where the drivers of the phenomenon are too diverse. Local 
9
prevention agendas must evolve around specific prevention strategies, and working with diverse 
leadership is desirable. The focus should be on generating networks of specialized knowledge. As 
different systems may be contributing to the same development goal, future programming should 
also facilitate communication channels between specialized systems. Additionally, concentrated
efforts may also decrease the number of conduit organizations increasing bureaucratic efficiency 
as well reducing cost.
Second, efforts should focus on catalyzing a wider range of models for at-risk youth by giving 
clear tools to local actors to catalyze new models and enable the conditions for specific and 
promising innovations to be able to thrive. Future programming should invest in developing and 
adapting intervention models for which there is a strong evidence base, such as multidimensional 
therapy, addiction treatment, and psychiatric approaches. Programs need be fostering innovations
based on the concentration, coordination, and proactivity principles.
Third, it is imperative that future initiatives carefully consider the need to adapt programs. Even 
within a country, local contexts differ vastly. An “adapt-pilot-scale” process is strongly 
encouraged. It is important to verify if the conditions are conducive for the implementation to 
be carried out as desired. If the local conditions allow for adoption of the essential components 
of the theory of change, then the nonessential components should be adapted to the local 
context. If the essential components cannot be held under the new circumstance, then other
policy alternatives should be explored. Organizations should also be given ample time to polish 
implementation models to achieve successful transfers, and rigorous evaluations should be carried 
out. Investing in a few strong pilots might bring a higher payoff than scaling up suboptimal models. 
Finally, engaging with government agencies to create specialized capacity for the design and 
implementation of specific services aimed at reducing recidivism is of the highest importance. 
Future programming should support public institutions shift away from a public safety and punitive 
approach to a public health perspective. This approach should be based on the understanding of 
youth protective and risk factors. Law enforcement and judicial prosecutors need to work 
collaboratively with service providers, such as: health and mental health services, education and 
employment support, and family counsel, to facilitate the long-term recovery of offenders. 
Working with these institutions to further this approach is a necessary condition for successful
and sustainable prevention policies to thrive.
10
I. EVALUATION PURPOSE 
The purpose of this performance evaluation is to gauge the effectiveness of the USAID JPV 
program in promoting reductions in levels of crime and violence (accountability) and gather 
recommendations for improved future programming (learning). The evaluation
1. gauges the extent to which the USAID/Mexico crime-and-violence-prevention 
programming has contributed to USAID/Mexico’s Development Objective 1 (DO 1): 
crime and violence in targeted communities sustainably decreased
2. assesses the effectiveness of the USAID/Mexico crime-and-violence-prevention portfolio’s 
JPV activity in achieving DO 1 intermediate results: 1.1) at-risk youth likelihood to 
participate in crime and violence reduced and 1.2) LPSs strengthened
3. provides recommendations on more impactful, proven, and sustained approaches for 
future crime-and-violence-prevention programs
The key evaluation questions addressed in this performance evaluation are highlighted in the 
Methodology sections below. The primary intended audience for the evaluation findings and 
recommendations is USAID in Mexico and the Mexican government, although the evaluation 
team considers that these findings are also relevant to share with stakeholders who collaborated 
with USAID for the implementation of this program and that the lessons learned and 
recommendations can be widely shared with the public to help inform future crime-and-violence￾prevention policies. 
This evaluation was commissioned to the Center on Conflict and Development (ConDev) at 
Texas A&M University. ConDev seeks to improve the effectiveness of development programs 
and policies for conflict-affected and fragile countries through multidisciplinary research and 
education. Our research is focused largely on mitigating conflict related to food security, natural 
resources, youth populations, and local institutions.
11
II. PROGRAM BACKGROUND 
Mexico has experienced significant increases in crime and violence in recent years, adversely 
affecting the country’s security and economic growth, with an especially detrimental effect on 
youth. Much of the crime and violence is driven by criminal organizations battling to fill the 
vacuum left by the fracturing of large cartels. Violence and rapid urbanization, particularly in 
Mexico’s northern border cities, have contributed to increased crime rates and recruitment of 
youth into criminal organizations, as well as threatened the security and well-being of citizens. 
USAID, through the JPV activity, aims to contribute to the reduction of crime and violence in 
Mexico by strengthening the capacity of local crime-and-violence-prevention systems to design, 
implement, and evaluate prevention practices and policies.1
The USAID JPV activity, implemented by Chemonics International, Inc., represents the largest 
activity within the DO 1 portfolio and aims to strengthen the capacity and sustainability of local 
violence prevention systems for sustained results by improving the crime-and-violence￾prevention policies of the Mexican government at federal, state, and local levels. This activity was 
implemented from October 1, 2016 to September 30, 2020 for a total cost of $24,466,087.
JPV aims to contribute to the reduction of crime and violence in Mexico through the achievement 
of three intermediate outcomes, as reflected in the program’s results framework in Annex 1: 1) 
contribute to the improvement of the Mexican government’s crime-and-violence-prevention 
policy; 2) strengthen the main actors that make up diverse local systems of prevention; and 3) 
promote the adaptation of evidence-based crime-and-violence-prevention practices and policies 
within local systems. This results framework was defined in 2017 after USAID/Mexico requested 
a change in the program’s programmatic approach to reflect the local systems framework. 2 This 
framework is rooted in the understanding that achieving any development outcome depends on 
the contributions of multiple interconnected actors and that sustaining development outcomes 
depends on the sustainability of the local system, understood as its capacity to adapt to changing 
circumstances. Under this framework, USAID’s role should be one of facilitator instead of 
implementer. In response, JPV sought to build local ownership by providing technical assistance 
to enable local stakeholders to drive their prevention agenda and to fulfill the functions of a local 
system. To this end, JPV prioritized strengthening local authorities’ commitment to tackle crime 
and violence, increasing the programming capacity of CSOs, and continuing engagement of the 
private sector in violence prevention. The program is premised under the following theory of 
change: “By strengthening local systems, local actors will lead evidence-based prevention 
interventions. These interventions will target the communities and youth most at-risk. Through 
a combination of interventions to address individual, family, and community risk factors, as well 
as interrupt the cycle of violence, there will be a sustainable decrease in crime and violence.” JPV 
encourages local partners to focus on secondary3 and tertiary4 activities toward reductions in 
violence levels. 
1 This is the context as presented by USAID. 2 See “Local Systems: A Framework for Supporting Sustained Development.” USAID, April, 2014. 3 Secondary prevention focuses on at-risk youth. 4 Tertiary prevention focuses on youth who are jailed or on probation.
12
The activity was implemented in nine municipalities in five states: 1) Tijuana in Baja California, 2) 
Chihuahua and Ciudad Juárez in Chihuahua, 3) Escobedo, Guadalupe, and Monterrey in Nuevo 
León, 4) Zapopan and Tonalá in Jalisco, and 5) Morelia in Michoacán.
13
III. EVALUATION METHODS AND 
LIMITATIONS 
To answer the key evaluation questions of this performance evaluation, we pursued a two-stage 
approach following the program’s theory of change. A theory of change explains how the activities 
undertaken by an intervention contribute to a chain of results that lead to the intended or 
observed impacts. Following this logical order, in Stage 1, we analyze the results achieved in 
strengthening local systems. We identify implemented prevention interventions, as well as the 
local actors and beneficiaries who were supported by JPV. In Stage 2, we study whether the 
interventions, as implemented by the local systems, may have contributed to a reduction in the 
likelihood of youth engaging in crime and violence and to an overall reduction in crime and 
violence in targeted municipalities. The two-stage approach is presented in Figure 1, and below 
we present the methods used to answer each of the key evaluation questions. The results 
presented for Stage 2 are preliminary, because two components of the data collection, focus 
group interviews and youth survey, have been delayed due to COVID-19 and will be completed 
upon a safer environment adhering to USAID’s “No Harm Done Principle.” We discuss this in 
more detail below. 
Figure 1. Two-stage Methodology
III.1 STAGE 1 METHODOLOGY
In this stage, we focus on analyzing the program design and implementation and on understanding 
the results achieved on intermediate result 1.2 (LPSs strengthened). We answer the following 
key evaluation questions: 
14
1. To what extent was the design of the JPV program relevant and sufficient for achieving 
the stated results and contributing to reduced crime and violence in the targeted 
communities?
a. To what extent did JPV’s updated theory of change (i.e., targeting high-risk youth 
and local systems approach) hold true?
b. What JPV program elements were the most effective and least effective at 
promoting reduction in crime and violence in targeted communities?
2. What evidence is there that JPV programming has strengthened LPSs in a sustainable 
manner?
a. To what extent are the local actors capable of maintaining the improved crime 
prevention policies, to the extent that there are improved policies, without USAID 
support?
b. Have key avenues for reform and a reduction in crime and violence in the targeted 
communities been addressed through JPV programming and implementation? Why 
or why not?
As stated by USAID,5 many final performance evaluations fall into outcome evaluations. This type 
of evaluation tends to consider results at all levels of an activity’s or project’s logical framework, 
that is, whether outputs were produced, to what degree the purpose was achieved, and whether 
any change in the status of the goal could be detected. This evaluation stage is concerned with 
understanding what outputs were produced, to what degree the purpose of strengthening LPSs 
was achieved, how programs were implemented, and whom they benefited—to then investigate 
whether these activities were able to promote change on the end goals of the program. 
To answer these questions, we performed an in-depth desk review of all relevant and available 
documentation provided by JPV and USAID, including planning and performance reports, 
assessments, and performance-monitoring data. With this documentation, we mapped JPV’s 
“pathway to change” to determine to what extent the theory of change as implemented held 
true, as well as to provide an analysis of the activities as implemented versus the program’s 
objectives. It is important to mention that JPV tracked 22 performance indicators related to the 
results framework, as well as additional nonobligatory indicators related to the end goals. We do
not study all of these indicators because that information is available in JPV’s performance reports. 
Rather, throughout the analysis, we bring to light specific results as they become relevant to our 
analysis. 
We also gathered primary data of the local-level actors who were supported by JPV through an 
online survey and semistructured interviews. Both of these instruments were designed based on 
the 5R framework, which highlights five key dimensions of a system: resources, roles,
relationships, rules, and results, to investigate how JPV interacted with local systems and to 
measure perceptions on key results.
6 Data collection instruments can be consulted in Annex II. 
5 “Performance Evaluation Designs.” USAID, 2019. 6 The 5R framework provides a scheme to analyze how the key dimensions of a system interact to produce desired outcomes, 
as follows. Human actors, both individuals and organizations, interact within a system assuming certain roles and within a network 
of various sets of relationships. The interactions depend on the resources actors have to produce results, and the process of 
transforming resources into results is governed by a set of rules. The 5R Framework in the Program Cycle. USAID, October 
2016.
15
Data from the only survey were collected from April 8 to May 4, 2020. The survey was sent to 
a list of 147 stakeholders selected in collaboration with JPV, with a response rate of 65%.
7
Respondent statistics are presented in Tables 1 and 2.
Table 1. Online Survey Respondents by 
Region
Region # of 
respondents
% of total
Tijuana 27 16
Chihuahua 6 6
Ciudad Juárez 25 24
Monterrey 22 21
Guadalajara 12 12
Morelia 12 12
Total 104
Table 2. Online Survey Respondents by 
Sector
Sector # of 
respondents
% of total
Local 
government
23 2
Federal/state 
government
15 16
Private sector 7 7
Civil society 51 49
Academia 6 6
Other 2 2
Total 104
The semistructured interviews were performed online from April 15 to May 30, 2020 through 
Zoom.8 The team invited a total of 57 key actors for interviews, including local anchor and 
strategic organizations, nonprofit organizations, governments, consultants, and the JPV team, with 
a response rate of 82%. In addition, nine interviews were carried out with field coordinators and 
six group interviews were conducted with program operators of selected implementing 
organizations. Table 3 and 4 present the distribution of performed interviews by region and 
sector. Data from all interviews were codified using Dedoose, a web-application for mixed￾methods research that allows codification of qualitative information. In total, the team codified 
1,561 excerpts from the interviews. 
Table 3. Semistructured 
Interviewees by Region
Region # of interviewees
Tijuana 6
Chihuahua 8
Ciudad Juárez 10
Monterrey 9
Guadalajara 13
Morelia 4
JPV Team 8
National* 4
Total 62
Table 4. Semistructured Interviewees by 
Sector
Sector # of 
interviewees
% of total
Local government 10 16
Federal/state 
government
1 2
Private sector 1 2
Civil 
society
Directors 17 27
Field 
coordinators
9 15
Program 
operators
6 9
Academia 2 3
JPV team 8 13
Other 8 13
Total 62
7 We have a total of 96 responses from the original list. We have eight additional responses. Some of these responses correspond 
to different government areas. Most likely, these are public institutions that registered the name differently or in less detail than 
what we had in the directory. Two additional responses belong to organizations for which two people instead of one responded 
to the survey. We decided to leave these responses because they have no significant effect on any of the performed analyses. 8 The original plan was to conduct these interviews face to face and to spend 2 to 3 days per region for this purpose, as well as 
to conduct focus groups. Unfortunately, given the COVID-19 pandemic, the Condev evaluation team not able to travel to the 
regions, but we were successful in transitioning to an online strategy to perform semistructured interviews.
16
To answer whether the program was able to strengthen local systems, we developed a general 
comparative analysis of the municipalities where JPV worked. Based on the 5R framework, in this 
analysis, the main features of how JPV worked in each municipality were highlighted to draw 
conclusions on potential features that might foster or hinder progress. 
To answer whether the program was able to influence key avenues for reform, we took a closer 
look at specific interventions supported by JPV to determine whether investments followed the 
existing evidence of what is effective, based on the USAID Abdul Latif Jameel Poverty Action Lab 
(J-PAL)–commissioned report, What Works to Prevent Violence Among Youth? For this purpose, we 
developed case studies based on the experience of six organizations purposely selected in 
collaboration with JPV and USAID based on their successes and program characteristics. For the 
case studies, we analyzed documentation, performed interviews with program operators, and will 
collect data from direct beneficiaries through focus groups. The sample provides information on 
programs that integrate either focalized interventions, community-based interventions 
(territorial), or both. It also includes both organizations that work with the juvenile justice system 
and with the civic courts. Five of these organizations received a grant, while one had a subcontract 
(ProSociedad). Of those that received a grant, one was directly from USAID, and the rest were
from JPV. Finally, to analyze how JPV leveraged its resources to transfer and scale up promising 
prevention interventions, we purposely selected three intervention models (ROLE, Mente Libre, 
and the CJHM in Morelia and Escobedo) as case studies. These models were selected based on 
their success and potential for adaptation and expansion. Table 5 summarizes the sample 
characteristics. 
Table 5. Case Study Selection and Characteristic Summary
Selection 
criteria
JPV grant/subcontract USAID grant Juvenile justice 
system
Civic 
justice
Model 
transfer
Focalized
Centro de Atención al Talento 
(CEDAT)
ProSociedad
Consejo 
Ciudadano
CEDAT
Consejo 
Ciudadano
CEDAT
ProSociedad
CEDAT
ProSociedad
Territorial
CASA
La Tenda di Cristo
Supera
CASA
La Tenda di Cristo
Supera
Supera
Institutional 
reform
CJHM Morelia
CJHM
Escobedo
III.2 STAGE 2 METHODOLOGY
In this stage, we focus on assessing the effectiveness of JPV in achieving intermediate result 1.1
(at-risk youth likelihood to participate in crime and violence reduced) and to gauge the extent to 
which the activity contributed to DO 1 (crime and violence in targeted communities sustainably 
decreased). We answer the following evaluation questions.
17
3. What evidence is there that JPV has reduced the likelihood of at-risk youth in targeted 
communities participating in crime and violence?
a. What evidence is there that at-risk youth reached with JPV programming will 
engage in alternative, positive pathways and steer away from involvement in crime 
and violence, including with TCOs?
b. Is JPV’s coverage or reach of at-risk youth in targeted communities (i.e., number 
of at-risk youth reached) sufficient to initiate a detectable reduction in youth 
involvement in crime and violence? Why or why not? 
c. To what extent have gender-based roles, including but not exclusive to the 
mothers of at-risk youth, affected the ability of the interventions to achieve their 
stated results (effectiveness)?
4. What evidence is there that the USAID/Mexico crime-and-violence-prevention 
programming through the JPV activity has contributed to achieving the goal of reducing 
crime and violence within the targeted JPV communities?
a. Where and what are examples of the JPV activity contributing to reduced levels 
of crime and violence in the targeted municipalities? 
b. To what extent have citizen perceptions of security increased in targeted 
municipalities?
The objective of this evaluation stage is to address outcomes and cause-and-effect questions to 
understand the contribution of USAID assistance to the outcomes of interest. Specifically, we are 
interested in understanding whether prevention interventions, as implemented by the local 
systems, may have contributed to a reduction in the likelihood of youth engaging in crime and 
violence and to an overall reduction in crime and violence in targeted municipalities. To answer 
the questions of interest of this performance evaluation, we pursued a mixed-methods approach 
based on standard case study techniques, including both qualitative and quantitative tools and
survey experiments to draw conclusions on what might have worked or not. The core of the 
evaluation relies on the assessment of observational data.
First, we analyze whether JPV’s coverage or reach of at-risk youth in targeted communities was 
sufficient to initiate a detectable reduction in youth involvement in crime and violence. To this 
end, we gathered, to the extent possible, data from beneficiary databases from supported 
interventions. Second, we will perform focus groups with direct beneficiaries of the case studies 
(see Section III.1) to understand which intervention components might have worked or not 
and why. We will make sure to incorporate youth who, based on the various characteristics of 
their environment, are more likely and less likely to respond well to the interventions selected. 
The possibility to reach these youth will depend on the collaboration of local organizations in 
supporting contact with them. In addition, deep consideration will be given to USAID’s “No Harm 
Done Principle.” Third, we will collect data of at-risk youth in communities with high homicide 
rates through an original survey to measure individual-level exposure to and perceptions of crime 
and violence. Our stratified sampling procedure will include youth in the targeted municipalities 
as well as youth from similar municipalities which remained outside the realm of JPV. A vignette 
experiment is embedded in the survey to measure the effect of hypothetical exposure to the 
intervention on subjects’ attitudes toward criminality. Vignette studies use short descriptions of 
18
hypothetical scenarios (vignettes) that are usually presented to respondents within surveys in 
order to elicit how their judgments about such scenarios affect outcomes of interest, often 
revealing their perceptions, values, or social norms. This survey will be informative of the 
potential mechanisms through which the intervention may have affected the outcomes of interest. 
Unfortunately, given the COVID-19 pandemic, the team has had to delay the field work. We will 
perform the focus groups and collect the youth survey upon a safer environment.
Finally, the we will perform descriptive statistical analysis of existing secondary data on crime and 
homicide rates, as well on citizen perceptions of security, to understand the context under which 
this program was implemented and the observable trends in treated municipalities and the rest 
of the country. One important limitation to highlight is that existing government surveys do not 
provide panel data representative of the local level. Therefore, this analysis is restricted to the 
level of specificity allowed by the existing data.
19
IV. FINDINGS AND
CONCLUSIONS 
In this section, we present the findings of our analysis, organized as follows. First, we answer the 
questions of whether JPV was able to strengthen LPSs and whether evidence-based informed 
interventions were implemented. To this end, we present an analysis of 1) JPV’s theory of change, 
2) JPV’s contribution to institutional capacity-building, 3) JPV’s contribution to the generation of 
multisector prevention agendas, 4) JPV’s contribution to the implementation of evidence-based 
prevention interventions, and 5) JPV’s contribution to the scale-up of effective models. Next, we 
answer whether JPV contributed to a reduction in the likelihood of at-risk youth participating in 
crime and violence activities. To this end, we present an analysis of 6) coverage and targeting and 
7) JPV-supported interventions results. As explained in the Methodology sections, this analysis is 
preliminary given the delays in field data collection due to the COVID-19 pandemic. 
In the analysis, we highlight which of the questions of interest are addressed by each section, and 
we provide summaries of the findings in boxes.
STAGE 1: WERE LPSs STRENGTHENED AND EVIDENCE-BASED INFORMED 
INTERVENTIONS IMPLEMENTED?
2. To what extent was the design of the JPV program relevant and sufficient for 
achieving the stated results and contributing to reduced crime and violence in 
the targeted communities?
a. To what extent did JPV’s updated theory of change (i.e., targeting high-risk 
youth and local systems approach) hold true?
b. What JPV program elements were the most effective and least effective at 
promoting reduction in crime and violence in targeted communities?
Box 1: Question 1 Summary
To answer this question, we analyze JPV’s theory of change, its design, and how it was implemented. 
We also review the program’s expenses to assess whether investments were directed to the activities 
as envisioned by the theory of change. 
Conclusion: JPV’s theory of change was successfully reflected in the strategy developed by Chemonics
International, Inc. and in the activities it envisioned to implement to achieve its goal of reducing crime 
and violence, based on working with local actors to fulfill three main functions of local systems: 1) create 
knowledge, 2) implement solutions, and 3) improve practices. The implementation commenced from 
the first function and gradually moving to the third. Because the practical implementation of JPV was 
launched early 2017, it did not have enough time to generate stronger improved practice outcomes. 
The program achieved stronger results in developing a shared understanding of the local violence 
phenomenon and in developing general organizational capacities at the local level, as well as moderate 
results in identifying promising practices and in implementing evidence-based models. Due to lack of 
evaluation of implemented solutions that are necessary to develop appropriate lessons and impact 
20
policy, improved practices were not as apparent. In the rest of the document, we review in more detail 
which components of each strategy were more and less effective.
IV.1 JPV THEORY OF CHANGE: THEORY AND PRACTICE 
JPV aims to contribute to the reduction of crime and violence in Mexico through the achievement 
of three intermediate outcomes, as reflected in the program’s results framework in Annex 1: 1) 
contribute to the improvement of the crime-and-violence-prevention policy of the Mexican 
government; 2) strengthen the main actors that make up diverse local systems of prevention; and 
3) promote the adaptation of evidence-based crime-and-violence-prevention practices and 
policies within local systems. As mentioned above, this results framework was defined in 2017 
after USAID/Mexico requested a change in the program’s programmatic approach to reflect the 
local systems framework. In response, JPV defined an LPS as a set of interconnected actors 
belonging to the public sector, academia, civil society, or private sector who work together to 
reduce crime and violence incidence by identifying the causes of violence and implementing 
specific solutions to address them, while also ensuring knowledge management of locally
generated learnings. JPV further defined that this system must generate individual incentives to 
encourage member participation and that incentives must be defined for other actors to join the 
system over time. 
To implement the local systems framework and reflect the theory of change, JPV developed a 
local systems strategy based on the three main axes that crime-and-violence-prevention 
systems must fulfill: 1) create knowledge by generating strategic information about crime and 
violence in the region, problems affecting the territory, types of violence, and types of affected 
populations under a prevention-and-risk-factor scheme, as well as identifying potential solutions; 
2) implement solutions by strengthening institutional capacities to design, implement, and 
monitor solutions that address the identified problems in the territory; and 3) improve 
practices by evaluating the implemented solutions with the objectives of applying lessons learned 
to the development of other solutions and impacting public policy. JPV sought to build local 
ownership by providing technical assistance to enable local stakeholders to drive their prevention 
agendas and to fulfill the functions needed on each axis. 
JPV focused its efforts on strengthening six LPSs. Some of these regions, Morelia, Escobedo, 
Monterrey, and Guadalupe, were previously selected to become part of the “city network” 
strategy pursued by JPV before the change in strategy. Seeking to expand its reach, JPV then 
added Tijuana, Ciudad Juárez, Chihuahua, Zapopan, Tonalá, and Guadalajara. Furthermore, given 
the new definition of local systems based on the interconnectivity of actors rather than on 
regional boundaries alone, JPV redefined Zapopan, Tonalá, and Guadalajara as the metropolitan 
area of Guadalajara and Guadalupe, Escobedo, and Monterrey as the metropolitan area of 
Monterrey. Throughout the document, we will refer to these metropolitan areas simply as 
Guadalajara and Monterrey.
Figure 2 shows the functions that comprise each axis of the local system strategy and the 
activities that JPV implemented to fulfill them, and it also displays the relationship of these 
activities with the program’s results framework. Given that JPV supports local actors in employing 
proven approaches and targeting the most at-risk youth for a sustained reduction in crime and 
21
violence, to better reflect the scope of the program, we added a fourth intermediate result to 
the results framework: 4) reduce the likelihood of youth engaging in violent activities. As the 
Activities row shows, JPV’s technical work was focused on providing and facilitating trainings, tools, 
and solutions to enhance institutional capacities and evidence use and generation for the 
implementation of crime-and-violence-prevention policies. Trainings were targeted at the 
government and civil society sectors, and tools were designed to engage with local actors to 
measure their capabilities and design specific strengthening plans. Most importantly, the program 
sought to transfer these tools to local and strategic actors to ensure sustainability. JPV generated 
partnerships and contracted services to implement solutions and to generate knowledge on 
specific solutions. It provided technical assistance to local institutions to implement these 
solutions and helped them systematize best practices. Finally, it pursued policy influence strategies 
to shape the public debate and originate policy changes within systems. 
Figure 2. JPV Local Systems Strategy and Implemented Activities in Relation to the 
Results Framework
Although not described as such by the program,9 we identify that JPV implemented a transversal 
institutional-strengthening strategy and four substantive transversal strategies: a) solutions for at￾risk youth, b) civic justice, c) gender strategy, and d) environmental design prevention strategy
(see Box 2). The last two are strategies that the program had initiated before the change to the 
local systems framework focused on at-risk youth. Environmental prevention became a low 
priority for the program, and we find that the gender strategy was managed in parallel to the local 
systems and at-risk youth strategies. Most of the analysis presented in this document pertains to 
9 The transversal strategies and strategic focus of the program changes across years. The evaluation team had to develop a 
framework in order to systematize the wide array of activities implemented by the program. We identified these as the main 
strategies implemented by the program throughout the years and organize all pieces of information accordingly. All the analysis 
is based on the three axes of the local functions and on these transversal strategies.
Development objective: Contribute to the reduction of violence and crime by strengthening adaptation and implementation capacities of 
local systems of prevention.
Intermediate Results
IR1: Support the Government of 
Mexico (GOM) in improving 
national violence and crime 
prevention policy and with its 
collaboration with local actors.
IR2: Improve the performance of 
actors within local violence and 
crime prevention systems.
IR3: Contribute to local systems 
generating and implementing 
evidence-based violence and 
crime prevention policy.
IR4: Reduce the likelihood of 
youth engaging in violent 
activities
System 
function
Activities
Generate knowledge
1) Build strategic alliances; 2) Generate 
local information about crime and 
violence; 3) Generate communication 
products; 4) Identify and disseminate 
good practices
Implement solutions
1) Create and agree on a common agenda; 2) 
Develop capacity to implement solutions; 3) Raise 
funds to implement solutions; 4) Implement 
solutions; 5) Involve relevant actors to implement 
solutions
Improve the practice
1) Evaluate solutions; 2) Establish 
transparency mechanisms; 3) Create a 
local training system; 4) influence public 
policy
- Trainings in C&V diagnosis 
and citizen participation
- Evidence reviews and 
initiatives
- Events and conferences
- Systematization of good 
practices
- Trainings in evaluation 
- YTT adaptations and users 
training
- YTT Platform
- Events and conferences
- Measurement tools: OPI-P, 
EBP, GOPI-P, YTT – & 
technical assistance
- Trainings in program 
design, Theory of Change, 
evidence use, monitoring, 
prevention strategy and 
program development 
- Transversal stratgies (Civic 
Justice, Gender violence, 
Youth at-risk, CPTED)
- Implementation guides 
and manuals
- Events and conferences
- Study tours
22
the institutional-strengthening, at-risk youth, and civic justice strategies, but we highlight results 
achieved by the other two when relevant. 
Box 2: JPV Transversal Strategies 
Institutional-strengthening strategy: Recognizing that “institutional strengthening and evidence 
generation for violence prevention interventions training is a quintessential factor that increases actors’ 
knowledge, leads to the successful implementation of prevention activities, and ensures the sustainability 
of the impact of these activities,”10 JPV implemented a mixture of learning strategies (both online and 
face-to-face) targeted at a wide array of actors from all sectors to improve their capabilities, including 
courses on strengthening violence prevention, strengthening institutional capacity, evidence generation, 
and the development of measurement instruments to be transferred to local actors.
Solutions for at-risk youth: Working with at-youth risk was the program’s priority. In order to 
develop local systems’ capacities to implement evidence-based prevention interventions, JPV developed 
a wide array of activities targeted at generating and systematizing evidence; identifying, strengthening, 
and systematizing effective interventions; sharing knowledge; and scaling up effective policies. With a 
particular focus on reintegration strategies, JPV implemented a donations program aimed at providing 
financial support to outstanding interventions while seeking to systematize their models and promote 
knowledge transfer. 
Civic justice: The CJHM was promoted by JPV based on the innovative approach of public officials in 
Morelia and Escobedo to deescalate violence and conflict and offer an alternative for everyday conflict 
resolution. This is an institutional model of community courts with restorative justice components, to 
which JPV also incorporated the “problem-oriented policing” (PoP) approach to strengthen the role of 
the police within the framework of civic justice. Through a series of technical assistance projects and in 
collaboration with the National Commission of Public Security, JPV focused on developing a 
systematized model that could then be transferred to other regions and developed trainings and 
materials to facilitate its adoption.
Gender strategy: This strategy was focused on family violence and sexual violence. The work on 
family violence was initially developed in the State of Chihuahua and was focused on the design and 
implementation of a care model for child and adolescent victims of family violence with a national and 
municipal focus. It then got transferred to Pachuca-Hidalgo, Guadalajara, and Monterrey. The work on 
sexual violence sought to build safe communities for girls, adolescents, and young women also in the 
State of Chihuahua. This strategy included the provision of specialized training though the Diploma on 
Comprehensive Care for Sexual Abuse during Childhood and Adolescence, developed by JPV in alliance 
with National Pedagogical University of Chihuahua. 
Crime prevention through environmental design (CPTED): This strategy, implemented in 
Chihuahua, Tijuana, Tonalá, and Guadalupe, aims to prevent crime and increase security perceptions in 
specific neighborhoods. Through this model, Government authorities, organized citizens, CSOs, 
academia, and the private sector work together to improve the environment through citizen 
participation activities such as the recovery of public spaces, cleaning of streets and areas of risks, social 
reinforcement activities in common areas, painting works, and reforestation, among others. JPV 
developed a specific strategy to promote synergies between the government and CSOs and between 
the CPTED model and specific youth risk factors. 
10 JPV Annual Work Plan 2017-2018.
23
JPV’s theory of change was successfully reflected in the strategy developed by Chemonics 
International, Inc. and in the activities it envisioned to implement to achieve its goal of reducing 
crime and violence. Figure 2 shows that all the intermediate results and their components are 
covered by the local systems strategy and its activities. 
The implementation of the local system strategy commenced from the first function and gradually 
moving to the third. Because the practical implementation of JPV was launched early 2017, it did 
not have enough time to generate stronger improved practice outcomes. This is confirmed by 
the LPS Performance Index (IRLSP being its acronym in Spanish) developed by JPV to measure 
the evolution of the local systems' three main functions. JPV applied this measurement instrument 
twice in 2019: one in January and February (reported in April), and a second one in September 
and October (reported in November). Table 6 presents the results (all scores are out of 100 
unless otherwise noted). The average results provide an overview of the areas where the 
program made more progress. The first axis, create knowledge, has the highest score, with 
averages of 71.33 in April and 72.35 in November. The second-highest scores are for the second 
axis, implement solutions, with averages of 62.13 and 70.4 (out of 125).11 Finally, the third axis, 
improve practices, shows the lowest scores, with averages of 36 and 43.09. As a general conclusion, 
this index shows that the local systems performed better in fulfilling the activities related to the 
first function, such as developing a shared understanding of local violence and building shared 
agendas, and it shows moderate performance implementing evidence-based interventions. 
Although the index shows an important improvement in the function of improve practices from 
April to November 2019, it depicts that the local systems still have ample areas of opportunity 
in related activities such as evaluating solutions and impacting public policy. 
Table 6. IRLSP (April and November 2019)
Region Axis 1. Create 
knowledge
(max. 100)12
Axis 2. Implement 
solutions (max. 125)13
Axis 3. Improve 
practices
(max. 100)14
Total
index 
(max 1.00)
Apr Nov Apr Nov Apr Nov Apr Nov
Chihuahua 74.05 75.36 79.16 77.03 44.58 54.17 0.61 0.64 
Ciudad 
Juárez
69.06 68.81 56.12 73.19 35 41.76 0.49 0.57 
Morelia 67.95 72.5 60.54 68.92 30.95 41.67 0.49 0.56 
Tijuana 69.02 76.61 60.79 73.05 36.72 45.36 0.51 0.6 
Guadalajara 61.09 69.30 44.52 68.53 32.55 41.33 0.43 0.55 
Monterrey 86.83 71.52 71.64 59.5 36.22 34.25 0.60 0.51 
Average 71.33 72.35 62.13 70.04 36.00 43.09 0.52 0.57 
11 This function also evaluates the adoption of a gender perspective in implemented solutions. The scores were low across 
municipalities.
12 The LPS generates and disseminates information about crime and violence, as well as the strategic problems that affect a 
territory, the main affected populations, and the most effective practices to prevent violence. 13 The LPS has the capacities to design, finance, and implement evidence-based solutions in response to the problem of 
violence. Furthermore, it encourages the creation of spaces and mechanisms for the articulation of actors from different 
sectors. This axis also considers the adoption of a gender perspective. 14 The LPS encourages the evaluation of implemented solutions. In addition, it generates training and strengthening mechanisms, 
as well as spaces for advocacy in public policy related to the prevention of crime and violence.
24
Stakeholders' perceptions also support this finding. We asked the online survey respondents to 
rate different statements about JPV’s contribution to each of the functions of the local systems
using a scale of 1 to 5, with 1 being totally agree and 5 being totally disagree.
● Regarding the create knowledge function, we asked stakeholders to rate JPV’s contribution 
to the following.
1. Identify, design, and implement programmatic high-potential prevention policies.
2. Local actors in the violence prevention system assign more resources to effective 
prevention strategies.
3. A diverse set of actors is more sensible about the importance of crime and violence 
prevention.
● Regarding the implement solutions function, we asked stakeholders to rate JPV’s 
contribution to the following.
1. Provide an elaborate diagnosis with disaggregated sociodemographic characteristics.
2. Have tools to identify and follow up with beneficiaries.
3. Improve capacity to monitor interventions.
● Regarding the improve practices function, we asked stakeholders to rate JPV’s contribution
to the following:
1. Organization has developing capacities to carry out monitoring and evaluation (M&E)
itself.
2. Facilitate cooperation with evaluation expert organizations.
3. The organization was able to influence policy during the last 4 years.
For the first and second functions, the results are quite positive, with averages of 1.82 and 1.86,
respectively. For the third function, the perception is less positive. For the first item, the average 
score is 2.02 and for the second is 2.25. Nevertheless, for these items, the anchor and strategic 
actors have significantly less positive impressions, with scores of 3.13 and 3.6 (leaning more 
toward disagreement). We deem that the difference in perceptions might be caused by the fact 
that anchor and strategic actors, overall, have more technical capacities than the rest of the 
sample, and, therefore, their vision is less optimistic. Also, there might be social desirability bias 
in these responses because actors self-report on their perceived evaluation capacities. 
Even though the improve practices function shows low scores in both the IRLSP and in our survey, 
and in particular on its evaluation component,15 from the comments we received from both the 
interviews and online survey, it is evident that JPV successfully managed to get organizations to 
internalize the need to have models informed by evidence and to evaluate their work. Also related 
to the third function, it is remarkable that 62% of stakeholders consider that their organization 
influenced policy (understood as the capacity to influence the decisions, actions, and policies 
adopted by the government). 
15 The component of evaluation of implemented solutions had the lowest score in the IRLSP of all functions in three of the six 
regions, with scores between 7.03 and 12.5 of 25. 
25
IV.1.1 JPV EXPENSES
To better understand the implementation strategy pursued by JPV, it is worth looking at its 
expenses. By providing financial support and facilitating and sponsoring specialized trainings and 
consultancy services, JPV sought to promote the rise and expansion of effective interventions 
that could be escalated or replicated in different parts of Mexico. JPV implemented different 
instruments to inject financial resources into the systems: a donations program (grants) and varied
subcontract instruments (subcontracts, purchase orders, and indefinite delivery–indefinite 
quantity [IDIQ]). Table 7 presents the distribution of expended resources, confirming the 
gradual implementation from the first to the third functions of the local systems discussed above. 
As can be observed, JPV assigned almost 60% of its resources to the first axis of create knowledge, 
40% to the second axis of implement solutions, and less than 1% to improve practices. This figure 
only shows resources assigned directly to evaluation technical assistance or projects related to 
the third function, but other contributions were made in this regard. 
In terms of transversal strategies, the biggest proportion, 39%, was spent on supporting 
interventions for at-risk youth. The implementation of the CJHM received 15% of the investment. 
Of these resources, about half were assigned to matters of training and institutional reform for 
government implementation, while the other half was assigned to create therapeutic solutions to 
channel youth through this system. The gender strategy received 10% of the expenditures, and 
CPTED received about 1%. Of the total, 35% was spent on the general institutional-strengthening 
strategy. 
Table 7. JPV Expenditures16 (All Amounts are in MXN)
Knowledge 
generation
(technical 
assistance and
training)
Implementation 
of solutions
Improve 
practices
(evaluation)
Total % of total 
expenditures
At-risk youth 10,247,763 24,100,941 295,800 34,644,505 39%
Civic justice 6,444,638 6,807,69017 - 13,252,328 15%
Gender strategy 4,580,019 3,953,601 - 8,533,621 10%
CPTED 350,000 610,250 - 960,350 1%
Institutional￾strengthening 
strategy
30,657,04018 - - - 35%
TOTAL 52,279,462 35,472,583 295,800 88,046,846
% of total 
expenditures
59.38% 40.29% 0.34 % 
Below, we analyze whether the implemented activities were able to achieve the expected result
of strengthening LPSs. For this purpose, we analyze in detail the quality and scope of the 
implemented activities and whether they were sufficient to observe improvements in each of the 
functions of the local systems.
16 These numbers exclude expenditure toward IDIQ that are still open, a total of 32,848,498 MXN, given that we do not have 
information on what exactly has or will be spent. Most of these resources are for capacity-building and knowledge generation. 17 These are interventions targeted at at-risk youth that are derived from the civic justice system. There is some overlap with 
interventions that were initially developed for at-risk youth in other settings. We only classify in this cell when the intervention 
or adaptation was specifically designed for the Civic Justice Model. 18 This number includes tool development activities, trainings, and overall capacity-building for local systems.
26
2. What evidence is there that JPV programming has strengthened LPSs in a 
sustainable manner?
a. To what extent are the local actors capable of maintaining the improved 
crime prevention policies, to the extent that there are improved policies,
without USAID support?
b. Have key avenues for reform and a reduction in crime and violence in the 
targeted communities been addressed through JPV programming and 
implementation? Why or why not?
Box 3: Question 2 Summary
For this question, we review the results achieved by JPV in strengthening local systems. First, in 
Sections IV.2 and IV.3, we analyze whether JPV was able to bolster local actors’ capabilities and 
promote the creation of multisector collaborations. Then, in Section IV.4, we analyze whether JPV 
contributed to the implementation of effective evidence-based policies by taking a closer look at the 
specific interventions implemented by local actors and their quality. Finally, in Section IV.5, we analyze
JPV’s contributions to the scale-up of effective interventions and the challenges involved. 
Conclusion: JPV achieved positive results with the implementation of the transversal institutional￾strengthening strategy, but the targeting of activities was too broad and involved a lot of local 
implementing partners. This policy had an opportunity cost of gaining stronger results with possibly 
fewer organizations implementing more focused programs. Even though the program had notable
success in increasing awareness and positioning the crime-and-violence-prevention agenda within the 
local and national systems, the results achieved in terms of generating sustainable, multisector,
coordinated agendas were mixed. JPV succeeded at enabling multisector collaboration when prevention
strategies aligned with interest groups, and government participation was obtained when related short￾term milestones were specified and followed. 
In terms of the implementation of evidence-based policies, we assess whether JPV supported 
interventions that followed the three principles of effective violence prevention found in the literature: 
concentration, coordination, and proactivity. We find that JPV technical assistance activities added to 
the concentration principle by supporting the development of civil society organization (CSO) programs 
with enhanced design, delivery, and coverage, although pressing challenges remain in terms of addressing 
youth risks and needs to prevent recidivism. JPV also contributed, but to a lesser extent, to the 
coordination principle by enhancing communication between CSOs and other relevant stakeholders, 
although major challenges prevailed with coordination with public institutions. The proactivity principle 
remains an area of opportunity for future USAID efforts. 
In terms of scaling up effective models, JPV was successful at leveraging different pathways to scale to 
expand two CBT models, ROLE and Mente Libre, and the CJHM. We identify that the focus from the 
beginning on scale, the provision of a meticulous strengthening process through access to overarching 
resources, and the generation of relationships of trust and co-investments with partners are key to 
achieving scale. Although JPV was very successful in supporting the adoption of these models, rushed 
expansion, absence of rigorous evaluations, weak technical capabilities in implementing organizations, 
and lack of consideration for differences in local circumstances posed challenges for model fidelity and 
effectiveness.
27
IV.2 JPV CONTRIBUTION TO INSTITUTIONAL CAPACITY-BUILDING
The institutional-strengthening strategy was the initial tool used by JPV to approach actors and 
one of the main resources brought into the local systems by the activity. Given this and the 
significant amount of resources invested in this line of action, we start the local systems analysis
by examining the results achieved by the mixture of capability-building efforts. The main 
achievement was increasing the awareness and positioning the prevention agenda within systems
and at the national level. This was accomplished through the development of capacities to include 
a prevention perspective into local interventions and by organizing conferences, events, and study 
tours, which facilitated knowledge-sharing within and between systems. However, we find that 
the targeting strategy of these capacity-generating activities was too broad, which came with an 
opportunity cost of gaining stronger results with possibly fewer organizations implementing more 
focused programs. We provide a detailed analysis of these findings below. 
Through the implementation of a series of trainings and development and transferring of 
measurement instruments, JPV was very effective at helping a wide array of organizations to adopt 
prevention programming by introducing concepts and methodologies such as theory of change,
evidence use for program design, risk factor diagnosis, and strategies for incorporating preventive 
approaches. Both the online survey and the semistructured interviews show stakeholder 
perceptions of these capacity-building tools to be quite positive. Table 8 shows that these tools 
successfully reached the majority of actors within local systems. It is remarkable that of all 
respondents, 84% know the trainings and 70% know the measurement instruments.19 As the 
numbers show, the usage of the tools is high. The most used tools are the trainings at 66% and 
the measurement instruments at 57%. Furthermore, of all respondents who indicated knowing
each tool, 86, 58, 52 and 41% consider the trainings, measurement instruments, study tours, and 
events to be useful, respectively.
Table 8. Knowledge and Use of JPV-Developed Tools
Measurement 
instruments 
developed by JPV 
(YTT, GOPI-P, OPI￾P, EBP)
Trainings Study tours Events/ 
seminars
Stakeholders who 
know the tool
70% 84% 20% 53%
Stakeholders who use 
the tool
57% 66% 8% 31%
Usage of those who 
know the tool
81% 78 % 40% 58%
Stakeholders who 
believe the tool is 
useful to design and 
implement better 
practices
58% 86% 52% 41%
19 The fact that only 20% of actors know the study tours is expected because this activity was only targeted toward specific 
actors. 
28
JVP developed four main measurement instruments, the Organization Performance Index—
Prevention (OPI-P) and the Best-Practices Scale (EBP being its acronym in Spanish) to work with 
CSOs, the Government Performance Index—Prevention (GOPI-P) to work with governments, 
and the YTT for general application (see Box 4 for a description of these tools). We find that 
the development of these tools is one of the greatest JPV contributions and that the provision of 
technical assistance, along with their application, impacted organizations in a positive way. 
Furthermore, JPV pursued a strategy to transfer these tools to local actors, which we regard as
outstanding because it allowed the program to reach more organizations while ensuring 
sustainability. Anchor and strategic actors agree that these instruments are technically sound and 
easy to use, which makes them accessible to a wide range of stakeholders. Below, we review the 
results achieved by the GOPI-P, OPI-P, and EBP. The YTT is analyzed in Section IV.4.
Work with governments for capacity-building started early in the program (2016–2017), which 
allowed for a medium-term perspective to implement capacity-building strategies. Annex III
presents the GOPI-P results across the life of the program. Overall, the municipalities supported 
by JPV increased their capacity by 50%,
20 which is remarkable. During the interviews, we also 
identified that the application of the GOPI-P instrument helps increasing awareness within diverse 
government institutions about the relevance of crime-and-violence-prevention work and about 
the need to coordinate prevention strategies. As one stakeholder expressed:
“…now the different institutions within the municipality know each other and know who 
does what. Previously, they were like islands. This would have been enough as an objective 
itself." 
From the interviews, it also became evident that the application of the tool, accompanied by 
continuous and reliable technical assistance, helped boost the government's capacities. Beyond 
the anchor actors in Chihuahua, Ciudad Juárez, and Guadalajara, who have fully adopted the 
implementation of GOPI-P, other local actors have also adopted the tool to reach more 
municipalities. Mesa Metrópoli, a strategic actor in Monterrey, has adopted the tool to implement 
in 13 additional municipalities of the metropolitan area, and the Jalisco State Center for Social 
Prevention has adopted the tool and will implement it in 80 municipalities. 
Our analysis shows that the work with the civil society sector was also very successful at 
increasing prevention awareness and generating capacities, although the targeting strategy for the 
provision of technical assistance appears to have been unclear for some tools. In collaboration 
with local anchor and strategic actors, JPV applied a total of 70 OPI-Ps to CSOs and provided 
technical assistance to a subset of them. As presented in Annex III, of 24 organizations that 
received technical assistance through application of the OPI-P, only 9 improved their score. Of 
24 organizations that applied the EBP tool and received technical assistance, 20 showed 
improvement. The EBP, unlike the OPI-P that covers organizational and technical capacities, is 
focused on the quality of evidence for specific prevention interventions to identify effective and 
scalable models. 
20 The municipalities that have had the greatest improvements are Monterrey (155 %), Ciudad Juárez (88%), and Chihuahua (86%). 
Conversely, Guadalupe and Tonalá showed a decrease in their capacity (–9 and –2%, respectively). JPV explains that this is caused 
by government changes.
29
Our analysis finds that increasing organizational capacities is an intensive endeavor, especially 
given the low existing levels as confirmed by the OPI-P results.
21 Therefore, working with 
organizations in narrower matters and with more focused interventions, as with the EBP, may be 
more effective. Moreover, the results from the EBP show that funding for a specific intervention, 
accompanied by close relationships and technical support, can really boost organizations’ 
capacities. All of the organizations that received funding and technical assistance through 
application of this tool improved their scores. In this regard, it is noticeable that only three of the 
organizations that applied the OPI-P received funding through the donations program, and only 
one organization has both the OPI-P and the EBP.
This broad targeting strategy is also reflected in the implementation of trainings. Of the online
survey respondents, 65% believe the content to be robust and 74% believe it to be relevant. 
However, we also identified a shared perception that, in order to achieve better results, trainings 
must be followed by more intensive assistance; otherwise, learnings are hard to internalize.22
Materializing training learnings into measurable changes within organizations is a medium- to long￾term process that requires more intensive work. It appears that JPV pursued a diversified strategy 
to work with organizations at various capacity levels and tried to reach several organizations 
within the LPS. Although the work to increase capacities is always valuable in itself, considering
the value for money of these investments is important. A broad targeting strategy raises questions 
about potential opportunity cost to having had more focalized efforts to achieve stronger results 
in the substantive transversal strategies.
It is important to reiterate that the evaluation does not find that the resources invested in 
capacity-building were misspent. The provision of organizational capacity-building activities is 
particularly important in the Mexican context, where often resources are lacking and baseline 
capacity levels are quite low. Furthermore, several organizations consider that the use of tools 
provided by JPV gives them legitimacy, not only because they expand their internal capacity to 
develop evidence that supports their work and strategies to communicate it, but also because 
these tools are implemented by external and expert entities, which gives them more credibility. 
As one interviewee stated:
“There is no longer an excuse to not ‘register’ models. Now it is possible to present our 
work to the government and businessmen with clear results thanks to the systematization
we have gone through. We are no longer only organizations with good intentions, we 
now have serious projects.” 
There is always a tradeoff between reaching more organizations versus strengthening a few while 
achieving stronger results. We believe that making this tradeoff transparent is important, 
particularly given the amount of resources that were assigned to institutional capacity-building. 
Overall, given the potential of the tools, we deem that more focalization could facilitate stronger 
results for the implementation of specific interventions. We discuss this in more detail in Section 
IV.4. 
21 In the last measurement of the OPI-P, only one organization was classified as high-level capacity, 11 have a moderate capacity, 
and 12 exhibit a low or basic capacity level. We include the baseline level of organizations that only have one measurement. 22 These numbers also include government-representative and private-sector perceptions about the trainings. 
30
Finally, JPV’s openness and willingness to discuss findings, have a dialogue, and adjust the technical 
assistance strategies according to organizational needs is noteworthy across tools. There is a 
general perception that the comments and feedback provided by JPV is valuable. More than 80%
of the stakeholders who answered the online survey expressed this as such. Furthermore, 73%
of stakeholders agreed that the modifications done to their programs or policies in response to 
JPV’s recommendations have helped them increase their effectiveness.
Box 4: GOPI-P, OPI-P, EBP, and YTT
The GOPI-P is aimed at measuring and evaluating the state of violence prevention capacity within 
municipal governments. Through the application of this tool, an institutional diagnosis is developed, 
identifying strengths and areas of opportunity. The results are delivered to municipal governments and 
include timely and feasible recommendations to increase capacities. The tool comprises five categories: 
1) cooperative government-society relationship, 2) evidence-based operational and administrative 
prevention tools, 3) services and specialized public servants, (4) municipal regulatory framework, and 
(5) information and communication systems. Based on the results, the GOPI-P categorizes government 
capacity in four categories: willing to participate (0–49), basic (50–74), acceptable (75–89), and superior 
(90–100). The GOPI-P was used by JPV to approach local governments since the beginning of the 
program and was applied in all municipalities where JPV worked. 
The OPI-P is aimed at measuring the institutional and technical capacities of CSOs that work on crime 
and violence prevention. Information gathered from this tool serves as a baseline for measuring changes 
in capabilities in the medium and long terms. Additionally, it supports the definition of strengthening 
plans that facilitate the organization's transition toward a results-based approach based on the 
generation of evidence. Through its variables, the tool promotes a theory of change that establishes 
adoption of internal changes and institutional policies that favor the organization’s performance in the 
field, positively impacting the places or people it tends to benefit. The tool evaluates 22 variables 
grouped under the following categories: governance, financial and administrative management, human 
resources, donations management, strategic planning, and technical capacity in violence prevention. It 
classifies organizations into four capacity groups: low, basic, moderate, and high. 
The EBP is aimed at identifying the use and generation of quality evidence in violence prevention 
programs with the ultimate goal of promoting evidence-informed policies. It analyzes the quality of 
evidence that backs up crime-and-violence-prevention interventions to identify programs that can be 
replicated or scaled up. The tool evaluates four categories: 1) theoretical background—analyzing 
whether the program has a design instrument and indicators, as well as whether it is based on existing 
evidence; 2) impact—analyzing whether the program has an evaluation, its quality, and results; 3) 
implementation quality—analyzing implementation performance and fidelity; and 4) alignment with 
prevention interventions that have been identified as effective in different contexts. It classifies 
interventions into four categories: not informed by evidence, informed by evidence, promising practice, 
best practice.
The YTT is aimed at assessing the risk profile of youth participating in CSO programs with three 
purposes: 1) ensuring that CSOs are focalized in high-risk youth, 2) having a baseline to determine 
individualized plans to intervene with youth, and 3) monitoring youth progress during participation in 
the program. The tool draws from the Structured Assessment of Violence Risk in Youth (SAVRY), along 
with recommendations from specialized Mexican institutions and organizations. Making use of open￾ended questions and a scoring scale, the YTT assesses the following: 1) overall risk of youth, 2) overall 
31
exposure to violence of youth, 3) propensity to perpetrate violence of youth, and 4) (when applied in 
group sessions) overall situation of a youth collective. The YTT implementation strategy relied on 
customizing it to CSO and public-institution programs that are directly working with youth and on 
training operators who, according to the tool manual, should be experienced in community processes 
with youth, providing psychological support to at-risk youth, working with youth in conflict with the 
law, and administering evaluations to determine youth needs. 
Beyond generating basic organizational capacities through the application of the measurement 
tools, JPV organized conferences and seminars with the aim of positioning the crime-and-violence￾prevention agenda in the public discussion, as well as creating awareness of evidence-based 
interventions and disseminating technical knowledge. The study tours, which were targeted at 
specific key actors, were organized to share experiences between regions and to facilitate 
knowledge transfer. Although only 41% of online survey respondents believe that the events were 
useful, from the interviews, we learned that the events and forums organized by JPV were 
successful in their objectives. First, they had an impact in generating pressure on governments to 
change the way they conceptualize crime-and-violence-prevention work and demand strategy 
changes. Second, they helped position different organizations’ work without atomizing the 
dialogues of individual demands. Third, they helped local organizations meet peers who
implement internationally recognized models, learn from their experiences, and face challenges. 
Our analysis finds that these events are a tangible way of bringing organizations closer to experts. 
A good example of this work was the event organized around CBT, which gave organizations the 
opportunity to hear from implementing experts in Chicago and Central America, as well as to 
learn from leading academics. As one stakeholder put it:
“Those events were incredibly helpful for us. We could explore different aspects and 
understand the CBT implementation perspective from leading organizations, such as the 
World Bank or universities. ”
As expected, there is more work to be done to have authorities commit to changes and the 
adoption of regulation modifications, but JPV’s approach to inclusive events in Mexico City and 
in each municipality helped advance the conversation. In Section IV.4 there is a more detailed
description of the technical content of these events. 
IV.3 JPV CONTRIBUTION TO THE GENERATION OF MULTISECTOR 
PREVENTION AGENDAS 
To implement the local systems framework, JPV worked to build multisector groups with shared 
crime and violence agendas within the regions and focused its efforts on identifying and working 
with diverse actors to enhance their prevention work. The program prioritized strengthening 
local authorities’ commitment to tackle crime and violence, increasing CSOs’ programming 
capacity, and continuing engagement of the private sector in violence prevention. Also, JPV opted 
to create anchor actors and strategic actors, with the objective of ensuring sustainability by 
empowering these actors to become articulators of collaborative efforts and to help other
members develop their capacities. In this section, we analyze JPV’s contribution to building
institutionalized multisector groups and how the program was able to influence key actors, their 
roles, and their relationships and to promote sustainability and local ownership.
32
We find that JPV enabled the coordination of sustainable multisector groups only when focused 
on the implementation of a specific prevention strategy. Relatedly, government participation was 
fostered when short-term milestones were provided. The strategy of having anchor actors was 
accompanied by several challenges that made its success limited, mainly given that the program 
failed in empowering local actors to lead the agendas, with relevant exceptions. In this regard, we 
find that the conceptualization of the anchor actor, while attractive, lacked appropriate 
coordination and application. Our analysis shows that this role did not exist or was not necessary 
in all systems and that working with diverse leadership is preferable.
We also find that JPV had a very interesting approach to involving the private sector in the 
prevention agendas, and although the co-investment of the sector was not considerable, JPV did 
manage to enable this sector to act as a key facilitator of local agendas. In terms of the civil society 
sector, JPV achieved remarkable results. It empowered the sector and placed it as a change￾generating actor by recognizing the innovative and programmatic value brought by CSOs to the 
systems, enhancing it, and creating spaces for organizations to showcase their work to the public 
and private sectors. Finally, we find that while JPV’s organizational structure with a regional team 
was highly effective in creating local buy-in and satisfaction by providing actors with local 
counterparts to discuss projects and ideas, the hierarchical structure of USAID, which reflected 
into JPV’s, negatively impacted local ownership. We discuss these findings in more detail below.
IV.3.1 MULTISECTOR GROUPS AND GOVERNMENT PARTICIPATION
As mentioned above, one of JPV’s greatest accomplishments was placing the crime-and-violence￾prevention agenda as a priority in the national and local discussions and articulating diverse actors 
to work together on the topic. However, as far as contributing to the development of 
institutionalized multisector groups with clear shared prevention agendas and working plans, the 
results are mixed. Even though JPV considers Chihuahua and Tijuana to be success cases, we find 
that the latter presents sustainability challenges. While efforts were initiated in Guadalajara and 
Ciudad Juárez, these didn’t start until 2019, so it is premature to talk about institutionalization. 
In Monterrey, several relevant actors were invited to the multisector group, but coordination 
remains a challenge. Finally, while in Morelia JPV did not pursue the strategy to develop a local 
system as in the other regions, an articulated multisector group has evolved around 
implementation of the CJHM. Annex IV presents a more nuanced analysis of the evolution of 
each system in a comparative matrix. Overall, we conclude that the most successful cases of 
multisector collaboration are those that have one of the following three attributes.
1) Groups that focused, at least in origin, on specific strategies rather than on 
developing broad crime-and-violence-prevention agendas. Our comparative analysis 
shows that articulation between actors happens faster and in a more natural way when articulated 
around a specific strategy. Morelia and Escobedo, with the implementation of the CJHM, are great 
examples, but so is the case of Chihuahua with the CPTED implementation. Moreover, our
analysis shows that groups focused on specific strategies rather than on developing broad crime￾and-violence-prevention agendas with ample menus, such as Tijuana and Monterrey, were able 
to better attract interest from the public and private sectors. As two anchor actors mentioned:
33
“The temptation in [crime and violence] prevention is to try to solve it all at once, and 
you end up solving little. It is better to give precision shoots.” 
“…it is necessary for the objectives to have a greater degree of focalization with a clear 
accompaniment so certain processes are easier to address.”
2) Groups that had short-term milestones attractive for government participation.
Mayoral terms last only 3 years. It is important to consider enough time for politicians to be able
to capitalize on successful interventions to get their buy-in to implement promising strategies. 
The prevention policy alternatives that can be implemented at the local level are not always clear, 
and public officials do not fully commit to their importance until they have the opportunity to get 
involved and understand their effects. Having champions within the government is a condition
that enables government buy-in, but it is not enough. The cases of Morelia, Escobedo, and 
Chihuahua show how JPV was able to leverage government interest by providing specific 
milestones on which politicians could capitalize. 
While changes in government are always a challenge, these two attributes together, specific 
strategies and short-term milestones, contribute to the probability of surviving administration 
changes. By showcasing achievements, the political costs of abandoning promising strategies are
high. The case of Morelia reveals important lessons in this regard. A clever communication 
strategy was used to convince the incoming government of the benefits of the CJHM, and a swift 
training plan was offered to keep the model in operation. The message was simple and clear, 
using anecdotes and available indicators, and there was a clear plan ahead. 
3) Groups that aligned agendas with specific interest groups. Our analysis shows that in 
bigger metropolitan areas, such as Monterrey and Guadalajara, or in areas with more complex 
crime and violence scenarios, such as Ciudad Juárez, organizing the crime-and-violence-prevention 
agenda under a unique group is not necessarily desirable or feasible. In more complex areas, there 
are several interest groups with diverse agendas, and, therefore, forcing them to collaborate 
under one umbrella may be impractical. As one actor put it:
“At the beginning the effort in Jalisco did not work because it started as an up-down 
collaboration trying to set an agenda that cannot be agreed. There should be a local 
vocation for things to work better. There should be incentives for people who are making 
decisions. That is more practical. Once that model is consolidated and you find the means, 
then it is possible to think about a public policy.” 
The results of these early efforts were groups without clear responsibilities and with coordination 
problems. For instance, JPV found it very challenging to develop an organized crime-and-violence￾prevention agenda in the Monterrey metropolitan area, and it was only able to articulate actors 
for the implementation of the CJHM through the last year of the program. The multisector work 
in Guadalajara only took off when organized around the implementation of the CJHM and 
reintegration strategies. In the case of Ciudad Juárez, different multisector groups have evolved 
around gender and social reintegration strategies. JPV conceptualized the systems work as a single 
group, but the experience shows that in complex agendas such as crime and violence prevention, 
a system can be composed of different groups that implement distinct strategies. For example, in 
34
a metropolitan area, civic justice will be more successful with buy-in from state authorities, while 
territorial strategies for at-risk youth may need stronger leadership from local authorities or 
CSOs. We deem that, in some cases, JPV faced an opportunity cost while trying to organize too 
many actors with different incentives under one agenda.
The takeaway from these attributes is that the prevention agenda cannot always be 
clustered within a single group and that specialization is desirable while facilitating
communication channels. The contribution of a program such as JPV is better suited to 
technically strengthen specific working groups and facilitate communication among them while 
working with different leadership entities and pursuing differentiated strategies. In sum, the above 
attributes mean that the construction of agendas must evolve around specific strategies with a 
clear technical offer, defined target populations and indicators, and clear responsibilities and goals.
One local system may have more than one agenda, and its construction must depart from clear
areas of specialization, such as at-risk youth, gender strategy, civic justice, or police work. In fact,
this is desirable. While there may be an overlap of actors among these groups, particularly from 
the government, evidence shows that it is more practical to coordinate different specialized 
working groups, where the government may participate in more than one, than to have a wide 
array of actors sitting at the same table trying to coordinate varied agendas. 
In terms of generating government buy-in, an important contribution of USAID is helping 
governments understand the value of working with at-risk youth and reintegration strategies. 
One of the biggest challenges faced by USAID at the local level is the lack of understanding and 
resistance of public officials toward a vision of prevention and reintegration, as these topics are 
still handled from a public safety rather than a public health perspective.
23 This agenda requires 
specialized and intensive services for specific groups, and aligning incentives with the government 
is particularly challenging because results are expected in the medium to long term and the 
number of beneficiaries is relatively low. The risk of not having specific interventions that can 
serve as an example for the government is that politician intentions remain in broad agreements 
that do not translate into operational changes. As several CSOs expressed during our interviews, 
the tangible and daily commitment of authorities, reflected in the will to invest in and facilitate 
specific interventions, should be the focus of this type of program. 
USAID developed a microeconomic model to estimate criminality costs. In Mexico, the economic 
and social costs of the criminal life of a young person exposed to different risk factors would 
total 27 million pesos. Considering a series of interventions with a hypothetical value of 2 million 
pesos, this report estimates the cost-benefit of investing in prevention and finds that for each 
peso invested in prevention, the long-term return can be 11 times greater.
24 The biggest 
opportunity is for governments to invest in this type of program. USAID can influence 
government priorities and promote co-investment when engaged with the municipality. 
Municipalities look for USAID prestige, and this should be leveraged as a vehicle to generate new 
ways to work with communities.
23 See “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018. 24 USAID, “Executive Summary: Estimation of the costs of crime and violence for the private sector in Mexico and the importance 
of its participation in violence prevention as a strategy for its long-term business model”. First Published June 2019.
35
IV.3.2 ANCHOR AND STRATEGIC ACTORS 
Following the evidence suggesting that strong and adaptive systems have actors playing roles as 
stewards, facilitators, brokers, knowledge hubs, networks, and advocates (5R framework), JPV 
decided to implement the roles of anchor actor and strategic actor as a key component of its local 
systems strategy. The objective was to ensure sustainability of their efforts, with these actors 
taking on the facilitator role in collaborative crime-and-violence-prevention efforts.
25
● Anchor actor is the actor who articulates the members of a system around a common 
goal or agenda, taking into account the roles, relationships, resources, and rules that 
operate within a system.
● Strategic actor strengthens capacities and functions of members of the system through 
the use of JPV-developed tools. 
The data gathered during this evaluation show that the strategy of having anchor actors was 
accompanied by several challenges that limited its success. While the experience varies by region,
the overall perception is skeptical. We identify four factors that impacted this result. First, 
contracts with anchor actors were not signed until late 2018, which left them with a very tight 
timeline, but with very ambitious goals. Second, we find that JPV did not take full advantage of the 
capabilities and knowledge had by local actors as prescribed by the local systems perspective. In 
general, anchor actors were not given the opportunity to lead the development of the local 
agendas. The mapping of relevant actors, decisions on whom to convene with, which institutions 
to offer technical assistance to, and how to build local agendas (thematic focuses of the working 
groups) were heavily influenced by JPV—and JPV’s perspective seemed to be above that of the 
local anchor actors. In terms of the strategic actors, we find similar perceptions. JPV, not the
strategic actors, decided which organizations were to receive technical support and for what 
purpose. In fact, these stakeholders shared that this selection often was at odds with local 
realities. From the interviews, it appears that JPV made decisions on which organizations to 
support based on the goals it wanted them to achieve, but not in terms of their baseline capacity 
or area of focus. The consequential pursuit of certain goals was, therefore, unrealistic in some 
cases and involved a high opportunity cost. 
Third, the rules that governed how the anchor and strategic actors took on their roles failed to 
empower local actors to lead the articulation of agendas. JPV offered subcontracts, which, by 
definition, are conceived by the sponsor and are relatively inflexible as to scope of work, budget, 
and other changes. Rather than providing financial support to achieve a public purpose, 
subcontracts are binding agreements between a buyer and a seller to provide goods and services 
in return for monetary compensation.26 Furthermore, these contracts have restrictions for the 
use of providers’ brands, which creates disincentives for sustainability and diminishes the 
leadership role. The role of subcontractor diminished the decision power of anchor and strategic 
actors to define strategy and invest in activities that they felt were the most promising and,
therefore, more sustainable. 
Points two and three are supported by data from the online survey. When asked about the 
distribution of decision power for program implementation on a scale from 0 to 100, with 0 
25 JPV’s Annual Report 2017-2018. 26 See Office of Research, “Federal Funding (Grants vs. Cooperative Agreements vs. Contracts)”, UC Santa Cruz. 
36
meaning more power concentrated in JPV and 100 meaning in the local organization, anchor and 
strategic actors reported an average of 37. Interestingly, the average for all other actors is 59. 
Furthermore, the online survey reveals that, indeed, local actors still perceive JPV/USAID as the 
articulation leader of the local prevention agenda. Of the total respondents, 73% believe that 
clear leadership exists to convene actors working on crime and violence prevention. Of these, 
39% believe this leadership to come from civil society, 17% believe it to come from the local 
government, and an important 35% still believe it to come from the international cooperation 
sector, understood as JPV/USAID. The perception of JPV as leader is particularly high for Ciudad 
Juárez, Chihuahua, and Tijuana, with 47, 40, and 39%, respectively. 
Fourth and finally, the selection of anchor actors may have forced organizations to take on several 
roles that may not be part of their nature. Some anchor actors, such as in Guadalajara, Monterrey, 
and Tijuana, were selected because of their role in the local systems, but not necessarily because 
of their technical expertise in prevention work, which may have complicated the evolution of the 
agenda. Moreover, from the interviews, we also find that other local actors questioned the 
designation of this role in some regions. In Guadalajara and Monterrey, the anchor actors have 
been more successful in institutional capacity-building (through the implementation of GOPI-P, 
OPI-P, and provision of technical assistance)27 than in the articulation of clear prevention 
strategies. The only case where the leadership appears natural was in Chihuahua and Ciudad
Juárez. Fideicomiso para la Competitividad y la Seguridad Ciudadana (FICOSEC) is a business 
organization with years of experience in security work and with funds to invest from a local trust. 
We believe this to be the only case where the anchor actor was able to fulfill its purpose. 
Whereas the case of Tijuana is generally considered a success, our analysis suggests caveats. 
While this actor thrived in bringing together a wide array of actors to work together on the 
prevention agenda, it was less successful in providing substantive technical leadership in 
prevention work. Moreover, given that this local system was not able to generate much co￾investment from the private or public sector, the sustainability of this role still seems to be highly 
subject to JPV. 
In short, the conceptualization of the anchor actor, while attractive, lacked appropriate 
coordination and application. Our analysis shows that this role did not exist or was not necessary 
in all systems. In more complex systems, such as metropolitan areas, there are several actors that 
play the role of articulators and leaders for different strategies. The existence or ability of one 
single actor who is able to articulate all the others seems to be at odds with the natural 
functioning of the local systems and may have the unintended consequence of creating power 
dynamics that contradict its objectives. As the 5R framework explains, the identification of the 
“to-be” system needs to be done in collaboration with local actors, and although there is little 
evidence, it hypothesizes that strengthening existing roles is probably more feasible than creating 
new roles.
28 We believe that certain cases would have been more successful if JPV had focused 
its efforts on strengthening diverse leaderships and leveraging its comparative advantages. In fact, 
as mentioned above, the evolution of the local agendas shows that they emanate from natural 
leadership and from organizations pursuing specific interventions. 
27 In Guadalajara, the anchor actor was responsible for implementing the GOPI-P and OPI-P and providing technical assistance. 
In Monterrey, the anchor actor was only responsible for the OPI-P and provision of technical assistance, whereas a different 
strategic actor was in charge of the GOPI-P and respective technical assistance. 28 See “The 5R Framework in the Program Cycle”, USAID, 2016 (p-13).
37
IV.3.4 PRIVATE SECTOR 
In Mexico, the philanthropy and social responsibility agendas are incipient and exhibit low levels 
of professionalization, making the involvement of this sector particularly challenging. 
Nevertheless, even though the co-investment of the sector in the violence and prevention agenda 
was not considerable, JPV did manage to enable this sector to act as a key facilitator of local
agendas. In particular, buy-in and support from different business organizations and companies 
were key to achieving the socialization of initiatives promoted by civil society and to contributing
to permanence of initiatives through government transitions. In this sense, JPV was able to build
a relationship with the private sector that enhanced its role within the system as an influential 
actor and increased its participation in the prevention agenda. The online survey shows that 54%
of actors believe the private sector to have increased its participation in this agenda.
The most successful case of private sector involvement and leveraging co-investment was in 
Chihuahua and Ciudad Juárez with FICOSEC, which, as mentioned above, was a natural choice
to also become the anchor actor. JPV succeeded in generating alignment with this actor and 
leveraged its resources, both financial and in terms of leadership, for advancement of the 
prevention agenda. In fact, the state of Chihuahua concentrates 68% of the co-investments able
to be leveraged by JPV from the sector, with a total of 5,777,907 MXN. From the total 8,540,318 
MXN leveraged by the program, the next-highest amount (20%) was leveraged in Mexico City. 
This means that only 13% of the total resources leveraged from this sector were from the rest 
of the regions. Nevertheless, JPV was successful in capitalizing on the private sector’s stature and 
enhancing its participation as an influential actor. For example, in Morelia, the Confederación
Patronal de la República Mexicana (Commission of Large Enterprises of the Mexican Republic)
(COPARMEX) took part in the civic justice system by offering some of its corporate social 
responsibility programs as alternative measures to justice. Then, this group also became an 
advocate, which facilitated the continuation of civic justice after the government transition. In 
Monterrey, Supera became an ally to advocate for at-risk youth interventions. In Tijuana, the 
sector became an advocate for increasing organizations’ visibility and strengthening community 
cohesion around violence prevention by generating convening opportunities. In one forum29
organized by JPV on the role of the private sector, one actor expressed:
“Decisions about prevention policies should not be only political, but should be sustained 
in evidence. The private sector can act as a critical mass to increase citizen participation 
in these processes.” 
One additional learning, and similar to the public sector engagement, generating interest from 
the private sector is more successful when involved in specific strategies. Such are the cases of 
Supera involved in an at-risk youth strategy in Monterrey, BBVA and Grupo Ruba involved in 
CPTED and gender strategy, respectively, in Tijuana, and COPARMEX involved in civic justice in 
Morelia. 
Finally, JPV supplied an important contribution in showcasing a model of local trust in Chihuahua 
as an avenue to direct resources to the prevention agenda in other regions. These efforts were 
particularly influential in Morelia, where a similar trust was created and discussions began 
29 USAID Mexico, Forum “The role of the private sector as part of the security and violence prevention strategy at the local 
level”, June 24, 2020. 
38
happening in other regions such as Guadalajara and Tijuana, although these efforts have not been 
materialized. In the same forum, one actor from the private sector in Morelia mentioned:
“We had the opportunity to know the models in Chihuahua, and thanks to this we had 
light on how to finance prevention efforts. USAID accompaniment is key to generate 
confidence to launch these types of trust models.” 
IV.3.5 CIVIL SOCIETY
In this portion of the analysis, we focus on understating how JPV empowered the role of the civil 
society sector overall, and in Section IV.4, we study in more detail JPV’s contribution in 
increasing its programmatic capacities. Through the local actor mappings, events, and multisector 
group discussions, JPV managed to give entrance to discussion from CSOs that otherwise find it 
hard to be heard. Moreover, JPV brought legitimacy to its work and placed them as change￾generating actors. JPV succeeded in recognizing the innovative and programmatic value that CSOs 
bring to the systems and enhanced their empowerment by facilitating the catalyzation of
innovative interventions with technical and financial support and by creating spaces for 
organizations to showcase their work to the public and private sectors. The online survey 
confirms this finding using a scale of 1 to 5, with 1 being low and 5 being high, on which 
stakeholders rated how positive they believe JPV’s influence to be on different sectors in helping
them improve their prevention practice. The highest rank was for civil society with 4.32, with an 
average of 3.62 for the rest of the sectors.
For all regions the highest rank among sectors in terms of JPV’s influence was for civil society, 
except in Morelia, where it was ranked third after academia and local government. This result is 
both expected and illustrative of the challenges faced by this region. While Morelia is a success 
story in terms of government buy-in, the civil society sector is quite debilitated. Box 5 presents 
a comparison between the regions of Morelia and Ciudad Juárez, which were the two extremes 
in terms of government and civil society empowerment, with Morelia being highly successful in 
working with the government but less so with the civil society sector and Ciudad Juárez being
the other way around. The case of Morelia shows that a successful local system is only such to 
the extent that it also involves strong CSOs, and this is particularly relevant for implementation
of the CJHM.
Box 5: Case Study—Ciudad Juárez and Morelia 
Comparative analysis of Ciudad Juárez and Morelia reveals important insights into the relevance of 
applying the local systems framework to specific strategies, particularly regarding the importance of 
multisector collaboration. In these two regions, JPV did not have a clear plan to build multisector groups 
from the beginning,30 but the agendas evolved naturally around specific strategies. These cases present 
advantages and challenges that are two sides of the same coin. In Ciudad Juárez, JPV was very successful 
at increasing CSO capacity to implement reintegration strategies, but lagged in creating more public 
services for at-risk youth. Conversely, Morelia is a success story in terms of government collaboration 
for implementation of the CJHM, but lagged in generating capacities for CSOs to collaborate with the 
government to support youth. 
30 From the interviews, we learned that Morelia was not considered a priority within the program until civic justice gained traction, 
and in Ciudad Juárez, there was not even a regional coordinator until 2018.
39
These cases show that focusing on a particular strategy is desirable to increase the delivery capacity of 
local actors, but also offer insights on the risks of not envisioning a clear coordination strategy between 
sectors to fulfill the strategy goals. In Ciudad Juárez, JPV focused its efforts on strengthening local CSOs 
working on reintegration strategies. This region received the biggest proportion of the donations 
program, which empowered CSOs to organize and form a Social Reintegration Committee. In 
collaboration with JPV and the anchor actor, this committee is working to influence public policy.31
Given the great work with CSOs, JPV leveraged 12,507,043 MXN in co-investments from the civil 
society sector in this region, representing 53% of the total leveraged from the sector across regions. 
Nevertheless, the program did not engage with the government to strengthen public services for at￾risk youth until late 2018, achieving poor results in this regard. This region received the lowest amount 
from public sector co-investment.32
The case of Morelia is a success story in terms of catalyzing institutional change through the 
implementation of the CJHM. It comes as no surprise that the co-investment from the government for 
this initiative was the second highest.33 However, there was very little work with CSOs, threatening
the effectiveness of referral processes of the CJHM and casting doubt on its sustainability.
34 The region 
received one-quarter of what Ciudad Juárez received from the donations program, and no resources 
where leveraged from this sector. 
JPV did implement strategies to address these challenges once the need to create synergies between 
the different sectors for the provision of specific services targeted at at-risk youth became clear, 
although these strategies were implemented toward the last one-third of the life of the program, leaving 
little time to achieve stronger results and ensure sustainability. 
IV.3.6 CONSULTANTS, THINK TANKS, AND ACADEMIC INSTITUTIONS
As part of implementing the local systems framework, JPV sought to gain substantive knowledge 
on specific topics from consultancies, think tanks, and academic institutions. The interviews 
revealed mixed experiences. Our analysis shows that smaller and locally based consultancy firms 
and research institutions that collaborated with JPV to develop specific tools had very positive 
experiences. In fact, beyond being grateful for the opportunities provided, they expressed that 
through JPV, they had a unique opportunity to influence policy. Nevertheless, the experience with 
external actors, bigger organizations, Mexico City–based organizations, and international 
organizations is less positive. Our analysis shows that, in these cases, JPV failed to take full 
advantage of their knowledge and expertise. Stakeholders’ perception is that these relationships 
were deemed as buyer-seller to cover one specific service requirement rather than being partner 
relationships building toward the construction of incremental agendas with iterative interactions. 
In particular, we find that organizations with established reputations and their own agendas found 
it difficult to collaborate with JPV. Two circumstances led to this result. First, there seemed to 
be an urgency to meet certain short-term goals, which led JPV to fail to consider expert advice 
31 Replicating an initiative that was born in Chihuahua and with the support of the local court, various government agencies and 
civil society actors participate in this interinstitutional working group and contribute to the social reintegration of adolescents in 
conflict with the law.
32 These minimal resources can be viewed in the comparative Municipalities Matrix in Annex IV. Moreover, these resources 
come from the Strengthening Security Program (FORTASEG being its acronym in Spanish), which is a federal subsidy for 
municipalities for security-related policies. 33 The highest co-investment is in Zapopan for the implementation of CPTED. 34 We will study the challenges involved in the implementation of the CJHM in Section IV.5.
40
or to let go of promising initiatives. Second, the co-existence of brands and intellectual property 
of the products developed by relevant actors is an additional challenge that emerged from the 
subcontracts used by JPV and that creates disincentives for continuity and appropriation. 
In this regard, we also find that, given USAID regulations, consultants cannot share which projects 
they are working on, hindering transparency and public discussions. As a consequence, the 
possibilities to leverage the knowledge of different firms is diminished. As one interviewee put it:
“USAID presents the results of reports without any type of peer review or opportunities 
for feedback, so there are no technical opportunities to improve ideas.” 
In order to truly advance cooperation, advance ownership, and build lasting relationships, 
activities such as JPV must relax these restrictions and allow actors to retain intellectual property 
of generated products. Otherwise, we find that the products will always be at odds with 
sustainability. 
IV.3.7 JPV ORGANIZATIONAL STRUCTURE
One of the greatest successes of JPV was the establishment of a regional team. From our 
interviews, we learned that most of the successful stories and satisfaction with the program derive 
from the work of this team. Stakeholders express the relevance of having this local counterpart 
to discuss ideas, concerns, and progress. Moreover, they highlight the openness of the staff to 
find solutions to any circumstance and their commitment to make their voices heard with the 
JPV central office and USAID. We received a wide variety of comments highlighting the technical 
preparedness and local knowledge of the regional staff, and the majority of stakeholders praised
them as one of the most relevant pieces to bring the local systems alive. Examples include:
“In our region, it was clear that the regional coordinator had a profound understanding 
of the issue and community to promote the agenda.”
“The communication with JPV was good, mainly through the regional coordinator who 
included us in all discussions.” 
“We expressed some concerns about the need to further adapt GOPI-P to our 
circumstances. The regional coordinator was always open to work in good harmony. 
Furthermore, although it was not always possible for him to be on top of everyone’s 
progress, he was always open to discuss feedback and recommendations with us.” 
One actor claimed that through this model:
“JPV found the key to make international cooperation sustainable.” 
One area of opportunity we find is the need to better align the technical work with the local 
systems team. These agendas seemed to have evolved in parallel at the beginning of the strategy
implementation, although they merged naturally as the program advanced. We learned that 
regional staff did not always take part in technical decisions, but JPV understood not to detract 
from a potential lack of very specific technical expertise when central teams lack awareness of 
41
the local context. In fact, we deem the success of several strategies to be attributable to the 
outstanding work of regional coordinators and to their previous professional relationships, which 
all of them leveraged to advance JPV’s agenda. For future projects, we deem that more 
empowerment of regional staff would supply potential for an even-more-local systems strategy, 
and we consider it imperative to give continuity to their work.
IV.3.8 USAID
As a USAID contractor, JPV decisions are restricted by USAID objectives, requests, and opinions. 
We find that one constraint for the application of the local systems framework is USAID’s 
hierarchical structure and dynamic with its contractors. The same logic of buyer-seller 
relationships that we described above exists within USAID. Contracts with activity implementers 
are regulated under acquisition procedures, through which the agency contracts out services to 
achieve certain desired outcomes. Under this arrangement, USAID typically exercises a higher 
level of control over the partner in obtaining results.35 While naturally some degree of supervision 
is needed, we find a tension between USAID’s management schemes and the enhancement of 
locally led development. 
First, USAID has direct control over high-level relationships at the local level. While involvement 
of high-level officials is desirable to further key partnerships, coordination and articulation are 
also needed with central and local actors. We found instances where JPV staff were restricted 
from having direct contact with or managing certain relationships, as well as instances where 
partners were requested to avoid direct contact with USAID because of their subcontractor role
with JPV. Second, USAID has other crime-and-violence-prevention efforts in the same regions 
and, in some cases, with the same actors who are managed outside of JPV’s local systems agenda. 
This is confusing for local stakeholders and prevents the consolidation of local systems. We find 
this case to be particularly relevant in Monterrey, where USAID directly manages high-level 
relationships with the private sector and leads parallel agendas such as Focused Deterrence or 
Red SumaRSE. Given the lack of clear communication mechanisms, we believe this to have
harmed the development of the agenda in this region. The existence of parallel relationships and 
agendas creates a forked agenda development, creating opportunity costs for the local system to 
leverage outputs and, in turn, to ensure intervention sustainability. 
The 5R framework reflects the principles of the collaborating, learning, and adapting (CLA) 
framework set forth by USAID to improve its development effectiveness.
36 One key aspect of 
this framework is the adapting piece, and evidence suggests that managing adaptively is more 
likely to improve outcomes when decision-making autonomy is placed as close to frontline staff 
and partners as possible.37 However, we find that given USAID’s hierarchical structure, the 
program’s decision making processes were highly bureaucratic and centralized. We find that 
oftentimes decisions had to be made in three different layers (USAID, JPV in Mexico City, and 
JPV regional staff and local actors), and that frequently, decisions were made in Mexico City.
Finally, we find that weak or nonexistent accountability loops from JPV and USAID to the actors 
within the system impact collaboration and sense of ownership. During the interviews, a 
35 See “Grant and Contract Process”, USAID, 2020. 36 See “Introduction to Collaborating, Learning and Adapting (CLA) in the Program Cycle”, USAID Learning Lab, 2019. 37 See “Evidence Base for Collaborating, Learning and Adapting. A summary of the literature review”, USAID, March 2020.
42
significant number of actors asked the evaluation team about the possibility to see the results 
from this evaluation and expressed general discomfort with being continuously asked to share 
their perspectives and experiences, but barely hearing back about the results from this type of 
exercise. Creating mechanisms for the international cooperation to be accountable to the local 
actors and share feedback is crucial when thinking about iterative collaborations and sustainability. 
When thinking about systems practice,38 it is imperative to recognize that USAID and its 
contractors become an actor within the system. By providing resources and defining the rules 
that accompany them, USAID creates a series of modifications in the relationships, interactions, 
and incentives that happen at the local level. USAID is not an external actor, but rather an active 
player in the system. Attributes of local systems are at odds with USAID’s hierarchical culture, 
and this culture reflects and has consequences outside of USAID. As such, USAID must address 
its own institutional barriers that diminish local ownership. While this is not particular to the 
Mexico mission,
39 we believe there to be important areas of opportunity to revise and reflect on 
these aspects to define a more inclusive and horizontal culture that better enables the 
development of local systems. 
IV.4 JPV CONTRIBUTION TO THE IMPLEMENTATION OF EVIDENCE-BASED 
PREVENTION INTERVENTIONS
In this section, we zoom into the specific interventions implemented by local actors to assess 
whether the program was able to influence key avenues for reform. After a short overview of 
JPV’s donations program, through case study analysis, this section examines the quality of JPV 
efforts to enhance CSO program effectiveness based on the three principles of effective crime￾and-violence-prevention strategies as found in the literature: concentration, coordination, and 
proactivity. The objective is to analyze the extent to which causal mechanisms could be theorized 
between the prevention interventions and the expected intermediate outcome of reducing the 
likelihood of youth to engage in crime and violence activities. That is, we analyze whether 
implemented interventions exhibit designs and an implementation quality that respond to what 
the evidence identifies as effective. 
We find most of JPV’s technical assistance activities to have added to the concentration principle 
by supporting CSOs program development with enhanced design, delivery, and coverage. To 
accomplish this, JPV improved the capacity of CSOs to use evidence, assess youth risks, specialize 
services through CBT protocols, and increase program coverage. However, pressing challenges 
remain in terms of addressing youth risks and needs to prevent recidivism. For this, more 
complex interventions with multidimensional strategies are necessary, and technical assistance to 
strengthen models with this approach is strongly recommended. In this view, work with CSOs 
with territorial interventions is a promising area. Additionally, an outstanding agenda is capacity￾building among government agencies to generate a specialized offer of youth services. 
38 See “The 5R Framework in the Program Cycle”, USAID, 2016. 39 USAID’s evidence review of CLA (2020) highlights certain attributes of learning organizations, such as flexibility in resources, 
risk-taking time, and flat organizational structure, that may be at odds with USAID’s existing culture for various reasons (p. 11). 
We believe the organizational structure to be particularly relevant for local systems development.
43
To a lesser extent, JPV contributed to the coordination principle. Main achievements here relate 
to enhanced communication between CSOs and other relevant stakeholders. This happened 
indirectly through knowledge-generation activities, which strengthened CSOs networks and 
facilitated communication with high-level government actors that may not be aware of CSOs
programs. JPV also helped CSOs navigate political changes and staff rotation in public agencies to 
ensure continuity in their collaborations with the government. Nonetheless, major challenges 
prevailed regarding CSOs coordination with the juvenile justice system and the civic courts, 
mostly related to youth referrals. Finally, the third principle remains as an ample area of 
opportunity for future USAID programs. While JPV did have lines of action to reduce recidivism, 
an important area of opportunity remains for programs that address underlying factors that cause 
violence. We discuss these findings in more detail below.
IV.4.1 KEY AVENUES FOR REFORM: WHAT THE EVIDENCE SAYS
When thinking about avenues for reform, the literature identifies overall key principles and 
elements of effectiveness that are crucial features of effective strategies and programs. Two of 
these principles are the concentration and coordination principles, which together establish that 
institutions should specialize in specific services targeted at the highest-risk places, individuals, 
and behaviors and should coordinate the provision of specific individual components with the
understanding of how different interventions may interact and complement each other. The 
literature proposes for institutions to focus on a narrow set of risks and protective factors in just 
one domain, but to coordinate a multisector response to address this complex multidimensional 
phenomenon.
40 A third principle is that of proactivity, which suggests that interventions should 
seek to prevent crime and violence before they take place by understanding and addressing the 
underlying factors causing them. Finally, this principle also suggests that, if crime and violence 
cannot be prevented, the focus should be on reducing recidivism by focusing on rehabilitation 
programs.
41
But designing a prevention policy and combining interventions are not minor tasks. The literature 
shows that certain interventions can produce strong and positive effects, but unfortunately, most 
of the interventions have mixed or small effect sizes.
42 Given the inexistence of “silver bullets” 
for crime prevention, strategies must contemplate the appropriate combination of modestly 
effective programs that, when combined, can produce robust effects. That is, prevention 
strategies most incorporate elements of existing effective programs while also promoting 
innovation based on key elements that underlie the effectiveness of successful interventions, and 
these interventions should pursue the concentration, coordination, and proactivity principles.
In terms of effective interventions targeted at at-risk youth, the literature shows that CBT￾informed interventions, multidimensional therapy, drug court, and drug addiction treatment, are 
effective interventions to deter at-risk youth from criminal and violent behaviors.43 These 
40 See Abt, 2017 and “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018. 41 See “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018. 42 Ibid 43 The literature also finds that focused deterrence and hot-spot policing are effective interventions. Focused deterrence programs 
have been effective in taking a narrow focus on deterring high-risk and chronic offenders from committing specific crimes (mainly 
homicide) by engaging the police and other service providers in providing targeted sanctions and rewards (J-PAL, p.10). We do 
44
interventions focus on understanding the drivers of individual behavior and shift related incentives 
or cognitive biases that may produce it, as well as support individuals with developing new 
prosocial skills. In particular, multidimensional interventions aim to prevent juvenile criminal 
activity by monitoring offenders inside and outside of school, furthering prosocial skill 
development, training caregivers and parents on behavior management techniques, and offering
substance abuse treatment. The USAID J-PAL–commissioned report finds that investing in these 
types of therapy models that can overcome weak family structures is of the highest relevance in 
the Mexican context. 
There is another group of interventions for which there is promising, but still incomplete, 
evidence. These interventions include vocational and training employment, restorative justice, 
alternatives to incarceration, and conditional cash transfers. The main caveats with vocational and 
employment programs are related to the fact that, while they are able to provide alternatives to 
crime, studies find that increased income alone may not always reduce criminal behavior. We 
know that there is a need to develop “softer” skills such as self-control and decision-making, but 
as the USAID J-PAL–commissioned report mentions, interventions also need approaches 
targeted at the underlying risk factors that affect, for example, gang involvement. In this sense, 
programs that are able to further nongang or noncriminal social identities or aspirations are 
desirable. Finally, there are a number of interventions with inconclusive evidence. These programs 
include community-based programs and school-based programs, among others. We encourage 
the reader to read the J-PAL–developed report, What Works to Prevent Violence Among Youth?, to 
find more nuances on the evidence findings.
IV.4.2 JPV DONATIONS PROGRAM
To implement the program’s prioritized strategy of catalyzing solutions for at-risk youth, JPV 
launched a donations program to identify outstanding interventions and, together with technical 
assistance, support their implementation, systematization, and evaluation. Also, the program 
organized a set of events to enhance knowledge-sharing and replication of successful 
interventions. After analyzing the evolution of the donations program and organized events, we 
conclude that the program evolved from a broader strategy looking at identifying promising 
interventions to supporting reintegration strategies to, finally, focusing almost exclusively on 
strengthening and transferring the CBT and civic justice models. 
Annex V presents the evolution of the donations program, showing the technical agenda as 
announced in the requests for proposal (RFPs), as well as the awarded grants. The first call for
proposals (2016–2017) was much broader and invited programs that had a variety of evidence￾based models.44 As this call was published before the change in strategy, it also financed programs 
outside of the six regions selected for the implementation of the local systems framework. As 
the program advanced during 2017 and 2018, the donations program took a much more focused 
strategy on the social reintegration of at-risk youth, while at the end (2019–2020), it pivoted 
toward financing projects to support at-risk youth as derived from the CJHM. Annex VI
not include this intervention in this analysis because focused deterrence is beyond the scope of JPV, and USAID manages this 
intervention separately from this program. 44 Including CBT, parental skills development, alcohol and drug abuse, restorative justice, social cohesion, new masculinities, social 
reinsertion, focused deterrence, gender-based violence (GBV), and environmental prevention
45
presents the events organized by JPV.
45 Early events were focused more on disseminating 
evidence on what works to prevent crime and violence and identifying promising practices (such 
as CBT). Then, the strategy concentrated on two main objectives: 1) sharing best practices for 
replication through study tours, mainly CBT and civic justice, and 2) positioning the CJHM and 
related interventions, such as community policing, on the national discussion and creating buy-in 
for their implementation.46
In line with the argument made above about the gradual implementation of the local system 
strategy, we find that JPV made good progress in identifying promising practices, but less progress 
in strengthening models to the point of generating evidence of effectiveness for replication in 
most cases. The donations program resulted in most of the interventions having systematized 
models that could later be leveraged by other organizations, which is an important contribution 
of the program. In fact, most of these learnings, together with other relevant publications from 
the program such as diagnosis and evidence reviews, are published in the Learning Virtual 
Community developed in collaboration with Tec de Monterrey. Nevertheless, as can be observed 
in the last column of the table in Annex V, only six of the grants have some type of evaluation, 
and most of these are results evaluations. Only CEDAT, Supera, Prosalud, and Desafío have 
slightly more rigorous evaluations, although most of them are still pre-post evaluations. From this 
table, it is also noticeable that only a few organizations received grant renewals, limiting the 
capacity of those that did not to further evaluate and scale up their models. In sum, we find that 
the evolution of JPV’s financial and technical offer, while having a clear strategy to scale up certain 
interventions, stopped short of generating evidence about the effectiveness of individual 
interventions and in furthering more programmatic collaborations for the implementation of at￾risk youth strategies that would reflect the concentration and coordination principles. Below, we 
take a closer look at the programmatic quality of the interventions implemented by local actors 
as a result of JPV support.
IV.4.3 PROGRAMMATIC QUALITY OF JPV-SUPPORTED INTERVENTIONS
Through analysis of the case studies, this section examines the quality of JPV efforts to enhance 
CSOs program effectiveness based on the three principles of effective violence prevention 
strategies: concentration, coordination, and proactivity. 
KEY PRINCIPLE 1: CONCENTRATION
From the three key principles of effective violence prevention programs, JPV’s technical assistance 
to CSOs mostly contributed to the concentration principle. This occurred through CSOs
program development making use of four mechanisms: 1) strengthened use of evidence, 2) 
enhanced assessment of risk factors, 3) specialization, and 4) program continuity and expansion. 
As shown in Table 9, all CSOs considered in this analysis experienced improvements associated 
with these technical assistance activities. 
45 In the text, we only discuss the events organized in the transversal strategies of solutions for at-risk youth and civic justice, but 
in the Annex, we include all of the events for the reader to see the evolution of the program. 46 JPV also invested in positioning the gender agenda in Chihuahua, but to a lesser extent.
46
Table 9. JPV Technical Assistance per CSO
Activity/strategy Territorial Focalized
JPV support/
CSO CASA La Tenda 
di Cristo Supera ProSociedad CEDAT Consejo 
Ciudadano
Use of evidence
None →
basic47
None →
basic48
Preexisting use 
of evidence49 +
focalization 
tool
None →
moderate50
Preexisting 
use of 
evidence51
Preexisting 
use of 
evidence52
Methodological 
strengthening
No support
provided
No support
provided
Gender-based 
violence
prevention 
(GBV) 
component 
and Youth 
Transforming 
Youth model 
development
ROLE model 
development
Model 
transfer and 
scaling
(incorporation 
of additional 
services)
No support
provided
Internalization 
of CBT and
mindfulness
No transfer Delivered 
among other 
activities
-Delivered 
among other 
activities
-Adapted 
protocol to 
include GBV 
and cultural 
activities
Focalization 
and fidelity 
vary among 
implementing 
partners
Core 
intervention 
(preexisting)
Delivered 
among 
other 
activities
Use of YTT and
monitoring 
tools
-YTT is 
used for 
JPV 
reporting
The 
intervention
also uses
additional 
tools to 
report to 
other 
funders
-YTT is used 
for JPV 
reporting, 
youth profile
generation, 
and socio￾demographic 
and legal 
interviews
-YTT is used 
for JPV 
reporting
The 
intervention
also uses an 
additional self￾designed tool 
for monitoring
-YTT is used 
for JPV 
reporting and 
youth profile 
generation 
and progress
tracking
The 
intervention
also uses 
additional 
tools for
monitoring
-JYTT is used 
for JPV 
reporting and 
youth profile
generation
The 
intervention
also uses 
additional 
tools for
monitoring 
tools
-JPV 
reporting
-Youth 
profile
generation
The 
intervention
also use 
additional 
tools for
monitoring
47 The program solidified its theory of change, intervention systematization, and development of strategies to enable replication, 
but weak M&E systems persist, as well as the absence of rigorous piloting. 48 The program solidified its theory of change, intervention systematization, and development of strategies to enable replication, 
but weak M&E systems persist, as well as the absence of rigorous piloting. 49 The program restructured the existing theoretical framework and added additions to existing procedures and tools. 50 The program developed manuals and procedures, M&E system to track youth progress, and rigorously piloted interventions, 
but challenges with the program’s theory of change, data and evaluation quality remain. 51 The program restructured the existing theoretical framework and added additions to existing procedures and tools. 52 The program restructured the existing theoretical framework and added additions to existing procedures and tools.
47
Strengthened use of evidence
This mechanism points to theory-of-change development, systematization of models and 
interventions, and strengthening of M&E systems. Based on CSO comments, these activities might 
have improved assistance and service delivery to youth. By developing the programs’ theory of 
change, CSOs enhanced their capacity to determine necessary interventions to address identified 
problems. As explained by two interviewees:
“CSOs, sometimes, sadly, work based on hunches. The theory of change helps 
understanding the causal logic of things. That is a major takeaway.” 
“…if we hadn’t understood the value of the theory of change, we wouldn’t know the 
causal relationship between our activities and where they are leading us…for example, a 
higher artistic knowledge leads to a minor drug use.”
Similarly, systematization of models and interventions strengthened the capacity of CSOs to learn 
from previous experiences and provided more resources to select strategies to work with new 
generations of youth. Even though it is not possible to develop manuals for every single 
intervention strategy and certain CSO models—particularly those following a territorial 
strategy—require some extent of flexibility to customize their work to specific identified needs, 
systematization did help them to have “more conceptual, methodological, and pedagogical 
clarity,” as pointed out by the director of another organization, and to replicate what has proven 
to be effective in the past. 
Although all CSOs did have M&E indicators before the JPV/USAID donation, only a few had actual 
M&E systems measured with reliable instruments that served to inform internal decision-making 
(see Table 9). M&E systems were enhanced or even entirely developed with the donations,
which helped CSOs have a clearer sense of intervention effectiveness and make more informed 
decisions on how to intervene with individual cases. As described by an interviewee from a CSO 
with no preexisting use of evidence on how M&E tools were developed:
“It was a trial and error. It is a deep exercise, a bit chaotic, not so organized…we built 
some [instruments] because USAID asked for a diagnostic…the other [tool] was also an 
exercise, because even if we have a vast experience, we had nothing organized nor 
systematized, we had to sit down and do the exercise to recover all that we knew to fill 
the registers, to do the diagnostics, the interpretative frameworks…now we go for larger 
groups, and that has pushed us to build the tools, profiles, models…before, it was 
improvised, now we have documents that support our work in the community…” 
Moreover, regular case monitoring with quantitative indicators, not only based on operator 
perceptions, serves to identify necessary changes to individualized intervention plans because it
informs on additional interventions or services that each youth may need to advance in their 
social reintegration process. These indicators are used, as well as a parameter to determine when 
someone is ready to leave the program, as the same interviewee explained: 
48
“These are long processes, the coldest part, the quantitative or measurable part, is when 
he has satisfied all considered variables in the program. Each program has a measurement 
tool. When he complies with all evaluation tools, [for example] when he has a practical 
or artistic skill, when he is emphatic with others’ problems…”
Overall, strengthening the use of evidence among CSOs is a clear achievement of JPV’s technical 
assistance. However, the extent of the internationalization of some of these efforts and their 
sustainability beyond the donation is not clear, particularly with regard to M&E systems and the 
use of JPV’s assessment tools, as explained below. 
Enhanced assessment of risk factors
A major part of JPV’s technical assistance to CSOs was directed to increase their capacity to 
assess youth risk factors, for which JPV developed the YTT (see Box 4) and encouraged its use 
and customization. The program required CSOs to assess youth risk factors making use of the 
YTT, which, beyond improving youth risk assessment and targeting, also supported CSOs reports 
to JPV on beneficiaries’ progress and results. Furthermore, it was customized to CSOs programs 
to increase its suitability. One interviewee explained the YTT customization process in his 
organization:
“We advanced a lot with JPV and the YTT. I think we destroyed the YTT, we did a new 
tool based on it. We have the same measurement logic, the same measurement method, 
but we cut half of the variables and introduced others. Actually, before starting the project 
with JPV, we used to do the profile based on our intuition and on information from the 
neighborhood…there wasn’t a clear systematization process and I think the YTT helped 
us building it…”
In the case of Supera, the YTT served as an input, along with preexisting tools, to develop an 
improved version of the organization’s own pre-post assessment tool with the support of JPV’s 
technical consultants: 
“The YTT determined the level of risk and the pretest determined there was an actual 
cognitive change in him. Because the YTT is not a pretest tool, we had to adapt it to the 
pre and post, and to intermediate evaluations to determine progress in them…The part 
of the posttest stems from the YTT, but we had already applied it in 2014 and 2015 with 
USAID’s support. What we did is using what we already had, along with JPV’s technical 
assistance, to build it — not just the YTT, which we have to apply because it’s the 
donation-- we were able to generate a more solid tool, that statistically can determine if 
there is a change and not only based on qualitative information.”
In spite of its technical value, it is not clear if the YTT will be internalized beyond JPV. On one 
hand, CSOs with preexisting use of evidence (see Table 9) already had tools for youth 
assessment at the different stages of their programs (eligibility, monitoring, follow-up), which, in 
the cases of Consejo Ciudadano and CEDAT, are more robust and specialized. There, the YTT 
has a low added value. In the case of SUPERA, it is expected that the newly developed tool will 
continue to be applied. In contrast, CSOs that had limited or no M&E systems before the donation
49
do recognize its value, but perceive the YTT as another administrative requirement. In the specific 
case of CASA, the tool seems to be co-existing with other instruments utilized to report to other 
funders, and its main function seems to be reporting. On the other hand, all CSOs, with the 
exception of ProSociedad—which is not applying the YTT—identified opportunity areas in its 
design and pointed out difficulties in its application. In general terms, there are three main areas 
for improvement:
1. Length: the questionnaire is too long and is hardly applicable in one single session—
according to one interviewee, a YTT session can take up to 3 hours. This is quite a 
relevant issue for organizations with a territorial intervention strategy that often identify 
their potential beneficiaries in high-risk areas (i.e., picaderos). Overall, the tool was 
designed to be administered in a private space, which does not correspond to the reality 
of CSOs with a territorial approach.
2. Bias and stigmatization: the tool is focused on the assessment of risk factors and exposure 
to violence. According to some interviewees, this generates a bias toward negative 
personal characteristics, which, unintendedly, stigmatizes youth during the interview. 
Furthermore, to have a complete assessment to develop an individualized intervention 
plan, it is necessary to address protective factors as well, which is what some CSOs did 
during the YTT customization process. 
3. Sensitivity: the questionnaire addresses highly sensitive topics for a first-contact interview, 
when it is too soon to have the necessary rapport and trust between youth and the 
operators, so the youth tend to lie. For instance, it is hardly probable that in a first 
interview someone recognizes or will admit to having a drug abuse problem. This implies 
that the initial profiles of youth are often inaccurate and have to be adjusted once the 
operators know them better. 
Additionally, it is worth noting that the YTT’s highest added value occurred where there were 
no assessment instruments before JPV’s donation. A clear example of this is civic courts, which 
did not have any tool to determine youth profiles and make decisions on referrals before adopting 
the YTT. However, training on how to apply the tool has to be reinforced because public officials 
lack specialized training to conduct assessments, as we will review in more detail in Section 
IV.5.
Specialization 
This third mechanism contributed to the concentration principle with efforts to enhance CSOs
specialization to work with at-risk youth. However, as discussed in more detail in this section, 
there is significant room for improvement to develop specialized services among analyzed CSOs.
To enhance CSOs specialization, JPV followed two different strategies at different stages in time, 
which are described in the above analysis of the donations program. In the first stage, the program 
conducted knowledge-generation activities: dissemination and study events on what works to 
prevent crime to identify best practices. In this process, CSOs learned from other CSOs, 
researchers, and relevant stakeholders by sharing their models, intervention strategies, and 
experiences. This might have increased CSOs awareness of extant tools and methodologies in 
50
the field, but questions remain on how these experiences were internalized by participants to 
improve their own programs. 
In a second stage, JPV focused its efforts on facilitating CBT protocols to CSOs. First, JPV focused 
its efforts on developing protocols: CEDAT’s Mente Libre and ProSociedad’s ROLE. In the case 
of ProSociedad, JPV funded the development of the CBT protocol for the ROLE model. 
CEDAT’s Mente Libre protocol already existed before the JPV donation, but JPV supported its 
strengthening and scale-up, with CEDAT’s role being to train other CSOs. Then, the program 
focused its efforts on scaling up these protocols. As shown in Table 9, from our case studies, 
JPV was successful in transferring the CBT tool to Consejo Ciudadano, Supera, and La Tenda di 
Cristo, which reported using it as part of their interventions. These protocols were also 
transferred to be applied as part of the CJHM. All of these transfers are analyzed in more detail
below. 
In addition to this, JPV supported professional development of CSOs staff and public officers 
through online training during the life of the program, resulting in certification in violence 
prevention and a course on violence prevention against adolescents. The certification, named 
“Violence Prevention and Strengthening of Citizen Security,” lasted 140 hours and was delivered 
in partnership with the Fundación Carlos Slim and the former Vice-Ministry of Prevention and
Citizen Participation. The course, named “Prevention of Violence against Adolescents and their 
Participation in Building Social Fabric,” was developed in conjunction with the National Human 
Rights Commission and is now part of the commission’s permanent online-training offer. Some 
CSOs interviewees, mainly directors and coordinators, mentioned having participated in these 
initiatives, although it is hard to determine how these trainings improved program specialization 
and delivery.
Overall, JPV’s efforts to build capacity among CSOs to address the highest-risk factors affecting 
most youth were limited. Specialized intervention in capacity-building was centered on facilitating 
the CBT intervention model. Although CBT has proven to be effective to produce temporary 
behavior change among young offenders, there are other community, family, and personal risk 
factors that need to be tackled to prevent recidivism. This is confirmed by the results of the YTT. 
As shown in Tables 12 and 13 in Section IV.6, where we analyze the YTT in more detail, the 
three highest-risk factors for men are community dynamics, drug abuse, and school context. For 
women, they are community dynamics, lack of a parental figure, and exposure to domestic 
violence. 
The urgency to address other risk factors is also confirmed by interviewee comments, who 
pointed out recurrent factors that interfere with youth adherence to programs and that hamper 
their sustained social reintegration. These factors relate to youth needs that are unmet by 
analyzed programs. The most salient ones are as follows.
● Drug abuse: Most CSOs do not have the technical capacity to work with youth who have 
drug abuse problems, in spite of the fact that this is one of the highest-risk factors 
identified by the YTT—particularly in the case of men. Furthermore, extant studies and 
evidence suggest that the benefits of interventions such as CBT might be null or limited 
for populations with substance abuse problems and that special protocols should be used 
51
instead.
53 Based on CSOs comments, two principal approaches seem to be in place to 
deal with this issue. The first one is not admitting them, but instead referring them to 
institutions specialized in drug abuse treatment. This approach is mainly taken by CSOs 
with a focalized strategy. The other approach is harm reduction, which operates in CSOs 
with a territorial strategy. 
● Family dynamics: Analyzed CSOs do not have major interventions with families beyond 
family therapy and orientation sessions in some cases. All CSOs operators highlighted the 
need to strengthen interventions in this area, as family is often the origin of youth 
antisocial behavior. However, JPV did not support work with families because these are 
indirect beneficiaries. This is an important gap in JPV and in CSOs programs because some 
of the most frequent risk factors relate to absence of a parental figure, domestic violence, 
and parents’ criminal activity, as shown by the YTT results. This was pointed out by several 
interviewees, for instance:
“We do work with them in several emotional aspects, but when they get home, 
there are violence issues that can affect them. In the end, the work does not have 
that much impact…they’re only 2 or 3 hours per week with us, but most of the 
time, they’re in their family context, thus there can be setbacks related to 
violence.”
● Lack of employment opportunities: Unemployment is a major threat to the social 
reintegration of youth after program conclusion, as economic necessities might push them 
back to criminal activity. In addition to this, at-risk youth tend to live in depressed areas 
where employment opportunities are limited and provide poor pay. As explained by one 
interviewee:
“…where youth live, it’s in the most distant areas from downtown, and job 
opportunities are 30 to 40 km from their communities. Thus, distance is even 
longer. We had already made some connections with youth, but they can’t go 
because with public transport, it would take them between 2 and 2.5 and a half 
hours to get there.”
CSOs have weak links with labor market intermediation services and insufficient connections 
with the private sector to link youth with job opportunities. Even Consejo Ciudadano, which has 
a strong relationship with FICOSEC, and Supera, which works with members from the private 
sector, face challenges in getting youth employed. It seems that for most youth, self-employment 
might be a better alternative, given barriers mentioned above and other factors, such as age (many 
are still minors), emotional instability, and stigmatization. La Tenda di Cristo and Supera
mentioned providing some support to youth entrepreneurial initiatives, although the use of 
specialized methodologies and expert advice in this field was not identified during the interviews. 
● Psychiatric follow-up: The need for psychiatric services was highlighted by only a few 
interviewees, particularly with regard to dual-disorder diagnostics and treatment, in spite
53 See Vaugh et al., 2003; Easton et al., 2007; and Friedman et al., 2008.
52
of the high sensitivity of this intervention area. Although mental health issues are common
among at-risk youth, none of the considered CSOs offers psychiatric services, and 
questions remain in terms of their capacity to diagnose mental disorders that would need 
more specialized treatment. 
Addressing these high-risk factors and youth needs requires multidimensional methodological 
approaches. In this view, a promising area of work is enhancing the specialization and intervention 
capacity of CSOs with territorial strategies, as their programs are more comprehensive and 
adaptive to the individual personal needs of youth, as illustrated in Table 10. Furthermore, these 
programs integrate strategies at the community level that can facilitate follow-up after program 
conclusion, which is a key aspect to prevent recidivism. 
Collected information from the interviews suggests that CSOs with this approach face more 
challenges and need more methodological strengthening to work in the different areas that they 
are addressing, along with enhanced process systematization and institutionalization. However, a 
major constraint possibly faced by JPV in bringing more support to programs with a territorial 
perspective is that intervention processes are longer than what was allowed by JPV’s regulations. 
Table 10. Comparison of Territorial and Focalized Model Strategies
Territorial Focalized
• Multidimensional approaches, but need 
methodological strengthening
• Customized interventions to youth 
characteristics
• Customized program duration
• Follow-up strategies after program conclusion, 
but informal and weak
• Lack of evaluation and systematization
(CASA and La Tenda di Cristo)
• Therapeutic protocols with supplementary 
activities
• Preestablished interventions with menu of 
services to be offered
• Preestablished program duration (judicial 
measures and protocols)
• No follow-up after program conclusion 
(Citizen Council for Security and Justice 
[CCSJ being its acronym in Spanish]: three
home visits)
• Manuals, evaluation, and systematization 
are integrated to operation
Lastly, an outstanding agenda related to the concentration principle—and one that was frequently 
pointed out during the interviews—is the lack of specialized government youth services. Youth 
are often sent to the standard public services offered, which often do not respond to their needs 
or realities. An example of this is public education programs, which do not consider constraints 
faced by youth who have been in a detention facility or who have been out of school for a 
significant amount of time. Organizations usually work with distance education systems, such as 
53
those offered by the National Institute for Adult Education (INEA being its acronym in Spanish)
or Prepa Abierta, but these do not ensure learning among these youth, and supplementary 
strategies have to be put in place with tutoring and remediation, as some CSOs have been doing.
This lack of public service specialization is reflected as well in public officer approaches—and even 
reluctance—to work with these youth. This was clearly explained by one interviewee:
“Institutions do not approach us to work…the kid is at the extreme of an institution that 
is not ready to work with youth…its rationale is not wanting to work with youth…there’s 
no institution that is willing to change and stop seeing youth as youth from the 80s or 70s, 
and to start seeing them instead as youth facing today’s problems.”
Program continuity and expansion
In general, donations helped sustain organizations’ work, as certain organizations did not have 
financial certainty to maintain their service provisions. Furthermore, donations also supported 
program expansion by increasing program services, which was possible thanks to JPV’s budgetary 
flexibility. This was clearly explained by one interviewee:
“We finally have a large team to implement the program. Also, the flexibility to have new 
items in the budget. There are funders that don’t support youth therapy, but USAID gave 
us the opportunity to have that, the same with medical services to prevent withdrawal 
symptoms, scholarships for training…to provide them with lunch.”
In most cases, USAID funding not only ensured the operational continuity of CSOs during the 
grant duration, but also enabled coverage expansion, either by being capable of reaching a higher
number of their target population or by opening up to new populations. An example of this is
organizations that were just working with youth in conflict with the law and that are now working 
with noncustodial sanctions and with civic courts. However, questions remain about the 
sustainability of CSOs’ work beyond JPV. This was pointed out by one director who highlighted 
the risks of only having one funder:
“I can’t imagine us without [USAID’s] technical assistance. I’ve got used to having this 
great ally…Fortunately, FICOSEC just approved one of our projects.”
CSOs also highlighted the value of USAID directing funds to work with these populations, which 
has had an effect in other funders that are now integrating these youth in their calls for proposals 
(i.e., Fundación del Empresariado Chihuahuense [FECHAC]). Furthermore, CSOs like La Tenda 
di Cristo and CASA mentioned the value of being capable of communicating and making use of 
evidence such as how they work and the effectiveness of their programs, which facilitate funding 
search processes.
54
KEY PRINCIPLE 2: COORDINATION
As explained before, the coordination principle points to mobilizing multisector responses to 
address violence. In this view, two of JPV’s strategies had some contribution to this purpose. The 
first one was knowledge-generation activities, which created connections between CSOs and 
government actors. They managed to enhance awareness and inform stakeholders on relevant 
topics on youth and best practices for intervention, but, more importantly, they helped to 
sensitize relevant local systems stakeholders on CSOs work and to improve CSOs coordination 
with public agencies, as explained by one judge who participated in these workshops:
“Two months ago, in Mexico City…the topic was boys, girls, and adolescents, I was the 
only one from the judicial branch there, the rest were CSOs participating in JPV…The 
training was on how to work with the judicial power, regarding CSOs and 
adolescents…these trainings are a substantial exchange of experiences on what civil 
society does in collaboration with authorities.”
JPV also supported CSOs in navigating political changes that accompany public officer rotation 
and often interrupt ongoing programs with government agencies. This was explained by the 
director of one CSO:
“JPV has helped us with the dialogue on some occasions, and in others, FICOSEC. That is 
one of the most important things that USAID should do: sit the stakeholders down and 
help grease the collaboration process…JPV’s way is politically correct because they have 
long-term relationships, thus we can have more private meetings [with the stakeholders]
where they can better understand the context or the programs’ guidelines, that the civil 
society can do it better because it has nothing to lose.”
Nonetheless, we identified major coordination challenges between CSOs and the justice system. 
These coordination problems relate to justice officers’ limited knowledge of services offered by
CSOs. This often results in young people being sent to CSOs that are incapable of assisting them 
and that have to redirect them, which interferes with their adherence to treatment programs 
and the social reintegration process. This issue appears to be particularly relevant for the CJHM, 
where youth are often sent to organizations that cannot even admit them based on their inclusion 
criteria, as we will review in more detail below. Other difficulties in the relationship with CSOs 
have to do with poor coordination among justice system officers, which can put youth at risk. 
Protocols need to ensure that public officers share information on the background of youth who 
are directed to the same organization and belong to confronted groups or were involved in the 
same crime. 
KEY PRINCIPLE 3: PROACTIVITY
Proactivity refers to interventions directed to prevent violence before it takes place, which entails 
addressing underlying factors causing violence. Although JPV supported a few programs with 
secondary prevention strategies, its focus was to work with youth in conflict with the law, so
most of its contribution to the proactivity principle was strengthening CSOs programs to prevent 
youth recidivism. As explained above, JPV’s technical assistance to CSOs did enhance their 
55
capacity to intervene with youth and to better address some of the risk factors affecting them. It 
is worth noting that, in the specific case of CSOs analyzed in this section, JPV did not support any 
community-based intervention or strategy intended to mitigate factors that can lead youth to get 
involved with violence.
The extent to which CSOs programs are capable of preventing recidivism is hardly measurable 
and comparable. On one hand, there are limitations in terms of data collection and information 
systems to track youth trajectories after program conclusion. Information systems of different 
states’ public agencies, such as public ministries, juvenile justice systems, criminal justice systems, 
and civic courts, are not sufficiently integrated. On the other hand, CSOs intervention models 
are different and target youth with varied characteristics. For instance, while certain CSOs do 
not admit youth with drug abuse problems, others do. Another example is the context of youth. 
While CSOs with a territorial approach are based in the highest-risk communities, CSOs with 
focalized models work with youth from different communities, which are not always the highest￾risk. This is important because community-level risk factors can push youth to recidivism. We 
present a targeted analysis in more detail in Section IV.6.
CSOs mentioned conducting follow-up visits or calls up to 6 months after program conclusion—
mostly as part of JPV’s requirements, although some had preexisting follow-up mechanisms. After 
that window, only CSOs with territorial strategies (see Table 10) seem to have informal contact 
and information about youth situations, as they perform regular activities in the communities 
where youth live. It is worth noting that interviewee perceptions on what constitutes effective 
social reintegration is varied. To some, their programs are effective in achieving social 
reintegration because youth do not come back. This was a more common view among 
interviewees from CSOs with focalized approaches. To others, social reintegration is a complex 
process with ups and downs that does not entirely depend on youth. This was explained by two 
interviewees from different CSOs: 
“Recidivism in criminal groups often relates to their economic situation. We have many 
kids that didn’t manage to transition to the labor market and ended up returning to 
criminal activity. What is more common is that someone in the family gets sick [and so
they start selling drugs again]. The families themselves push them to return to criminal 
activity [with economic pressure]. There are cases of mothers who poison their sons to 
avenge their brothers [when they are murdered].”
“The context, connections with organized crime, community, social, and family contexts, 
gangs’ presence. One in every three can relate to a psychological cause that was not 
addressed during childhood and to drug use. Working with friends is also necessary to 
enhance reintegration.” 
Considering this and as pointed out above, strengthening the programs of CSOs with territorial 
approaches is a promising area of work—not just because these are more comprehensive and 
adaptive programs addressing more risk factors and youth needs, but because they have sustained 
interventions in communities to enhance protective factors at that level. In the words of one 
interviewee:
56
“There are other factors, such as the community. We identify potential allies, support 
networks that can be mothers, certain police officers, school teachers, and even their 
own crew can constitute a protective factor.”
This is relevant from the perspective of the proactivity principle. Community-level protective 
factors can benefit all youth living there—not only those who have been through the juvenile 
justice system or the civic courts—and can help prevent others’ future involvement with violence.
IV.5 JPV’S CONTRIBUTION TO THE SCALE-UP OF EFFECTIVE MODELS 
This section analyzes JPV’s efforts to encourage the piloting, systematization, and scale-up of 
promising models to prevent crime and violence. We find that JPV was successful at leveraging 
different pathways to scale and that the focus from the beginning on scale, the provision of a 
meticulous strengthening process through access to overarching resources, and the generation 
of relationships of trust and co-investments with partners are key to achieving scale. Although 
JPV was remarkably successful in supporting the adoption and scale-up of the ROLE, Mente Libre, 
and civic justice models, rushed expansion, absence of rigorous evaluations, weak technical 
capabilities in implementing organizations, and lack of consideration for differences in local 
circumstances posed challenges for the fidelity and effectiveness of models.
JPV’s objective was to identify and enable the conditions for local systems to scale up effective 
policies. To this end, JPV allocated resources and provided technical assistance for the 
implementation, adaptation, and evaluation of interventions in the four substantive transversal 
strategies of the program and facilitated interventions’ systematization and knowledge transfer 
within and between regions to achieve scale-ups. JPV’s assistance consisted mainly of recruitment 
of specialized consultants, organization of trainings and events with key international and national 
actors, facilitation of knowledge transfer through study tours, publication of manuals and
implementing guides, funding of process and impact evaluations, and development of targeting 
tools. By providing access to overarching resources and focusing its technical assistance on policy 
influence and transfer, JPV was able to leverage its resources and interactions with the system to 
“close the cycle” and achieve the scale-up and transfer of promising models. The scale-up results 
achieved by each strategy are as follows.
• CPTED: Chihuahua’s municipal government implemented and evaluated a CPTED 
intervention, “My Neighborhood is My House,” which then was expanded to four 
additional municipalities. The CPTED methodology was also adopted in Guadalupe, 
Mexicali, Tijuana, and Zapopan.
• Gender strategy: The care model was transferred to institutions in Ciudad Juárez, 
Monterrey, and Guadalajara.
• Civic justice: JPV was able to influence the Mexican government to launch a national 
police and civic justice model in 2019, which incorporates in its entirety the Civic Justice 
Homologated Model and a PoP model. This model is being scaled up to more than 30 
municipalities in the country, with additional demand for replication in others. 
• At-risk youth: Two CBT-based interventions, ProSociedad’s ROLE and CEDAT’s Mente 
Libre, are being implemented by several organizations in Monterrey, Ciudad Juárez,
Chihuahua, Morelia, and Guadalajara and as part of the treatment services of the CJHM. 
57
Annex VII shows the resources and activities dedicated by JPV for each of these strategies and 
presents the results achieved in each function of the local system that led to scale-ups. In this 
section, we study the cases of CBT and civic justice because they achieved a higher degree of 
expansion, and we deem them to be the most successful. As reviewed before, these two 
strategies were a priority for the program and received significant investment. By studying these 
cases, we aim to shed light on the key resources and avenues that can contribute to successful 
expansion of the interventions, as well as to highlight remaining challenges. 
IV.5.1 PATHWAYS TO SCALE AND ELEMENTS OF EFFECTIVENESS
In general, evidence-based policy design consists of a constant process of adaptation and 
innovation, testing, and scaling-up. There are different ways in which implementing organizations, 
governments, and international development agencies can leverage evidence to inform and shape 
policy at scale. The key differences in the pathways to foster the transfer and expansion of 
promising interventions are the existing evidence bases from which these processes depart. JPV 
followed different pathways to catalyze, systematize, and scale the selected interventions.
• Adapting, testing, and scaling up an intervention model informed by intervention 
models of renowned evidence-informed programs. The ROLE model was developed 
by ProSociedad per JPV’s request. Given that there is a strong base of rigorous evidence 
supporting the effectiveness of specific CBT programs, such as Becoming a Man (BAM)
54
and Sustainable Transformation of Youth in Liberia (STYL),
55 JPV awarded a subcontract 
to ProSociedad with the objective of tailoring a CBT program for the Mexican context 
with components that could be transferred to other institutions and that could also be 
adopted in a school context. 
• Strengthening, testing, and replicating an innovative model informed by research 
insights of rigorous evaluations. The Mente Libre model was developed by CEDAT, 
and JPV facilitated its strengthening and expansion. Through a partnership with local 
academics, CEDAT had developed a model that accounted for insights from different 
disciplines such as psychology and neuroscience. JPV built upon these efforts and awarded 
two donations and a follow-up grant to CEDAT. The objective was to evaluate and 
systematize the model with a public policy approach in order to transfer it to the local 
government and other organizations.
• Innovating, systematizing, and replicating a promising institutional reform 
informed by elements of effectiveness. The Civic Justice Homologated Model is 
an institutional innovation based on community courts, restorative justice, and PoP
components that JPV developed in collaboration with the governments of Morelia and 
Escobedo and the federal government. Based on an idea that emerged at the local level 
and for which there was no evidence base or program to adapt from, JPV provided 
technical assistance to design a model, connected implementers with specialists in court 
innovation implementing community court models, generated administrative capacities for
54 See Heller et al., 2017. 55 See Blattman et al., 2017. 
58
its implementation, and systematized the model so that it could be transferred to more 
municipalities and influence policy at the national level.
We identified the following three common characteristics that made these scale-up models 
succeed.
1. JPV’s technical assistance focused from the beginning on developing 
interventions with potential for scale. Defining this clear objective allowed JPV to 
focus its efforts and resources on facilitating organizations’ access to platforms to 
showcase their work and on developing products that would give further exposure to the 
models. Additionally, because these products systematized the use and generation of 
evidence and provided tools for their implementation in other contexts, they were 
instrumental in gaining traction to drive their adoption and replication.
2. JPV implemented meticulous strengthening processes providing access to 
overarching human capital and financial resources. Access to overarching 
resources that contemplated the implementation, evaluation, and transfer of the programs 
and specialized technical assistance from consultants strengthened the models and 
generated buy-in from relevant actors in the different local systems. The access to 
overarching resources is reflected in the total amount of investment that JPV allocated to 
the strengthening, implementation, evaluation, and systematization of these models 
(approximately 20% of the total amount of the program’s expenses) and in the diversity 
of the services provided. Table 11 shows the contracts awarded to different 
organizations to achieve the scale-up of the selected intervention models in detail. 
Additionally, organizations such as CEDAT were among the few organizations that 
received multiple donations and even a follow-up grant to give continuity to the results 
obtained through the previous award. 
3. JPV was effective at building credibility and generating trust in the potential 
for expansion of the models, securing co-investments from implementing 
organizations and interest from relevant players. The credibility built by JPV and 
its responsiveness to local actors’ initiatives led CEDAT and the local governments of 
Morelia and Escobedo to designate sizable resources to strengthening and systematization 
and to invest significant efforts in facilitating model transfer (see Table 11). JPV was able 
to create trust relationships with partners. As one interviewee put it:
“The comparative advantage of USAID is that they say they will support evidence￾based programs to [scale up models], and to support evidence generation and 
provide funding for evaluations. This is something innovative that we had not seen 
in Mexico for 10 years. We will sit with you so you can evaluate your project, and 
if you do that right, we will give you the platform and we can refinance you…These 
secretly conditional grants are very good. I condition you to grow. Grow and I will 
give you more money.” 
59
Table 11. JPV Expenditures in the Scale-Up of Selected Intervention Models
Model Organizations involved in the 
scale-up of the model Total expenditure by type of contract and activity
ROLE ProSociedad (consulting organization
and developer of the model and 
transfer methodology)
Laboratorio de Estudios Económicos y 
Sociales (consulting organization and
evaluator of the model)
Total invested amount: 2,625,388.40 MXN
• Subcontract to ProSociedad to design and pilot the 
intervention model and develop a transference methodology to 
the school context: $1,729,176.80
• Purchase order to Laboratorio de Estudios Económicos y 
Sociales for the external evaluation of the program in public 
schools: $295,800
• Purchase order to ProSociedad to support the transfer of the 
model to organizations providing services to infractors in the 
civic justice process in Morelia: $600,411.60
Mente 
Libre
CEDAT (local nongovernmental 
organization [NGO] and developer and 
implementer of the model) 
Total invested amount: 3,125,048 MXN
• Grant to strengthen the model and assess the possibility of 
transferring the model to Jalisco’s detention centers: $999,598
• Follow-up grant to generate evidence of the impact of the 
model and transfer the model to Jalisco’s detention centers: 
$524,505
• Grant to design an integral program including Mente Libre as a 
component to serve youth in conflict with the law on 
noncustodial sanctions and/or civic justice processes in the 
metropolitan area of Guadalajara: $1,690,945
* Additionally, CEDAT co-invested 1,208,971.40 MXN
Civic 
Justice 
Homologa￾ted Model
C-230 Consultores (consulting 
organization, developer of the 
homologated version of the model, and 
advisor in the design of civic justice 
law)
Observatorio Ciudadano de Seguridad, 
Justicia y Legalidad (consulting 
organization and developer of training 
guides and sessions guided to public 
officials)
Inteligencia Pública en el Análisis 
Político (consulting organization and 
advisors and assistants to local 
governments in the implementation of
the model)
ProSociedad (consulting organization
and assistants in the adoption and
transference of the intervention model)
Universidad Vasco de Quiroga (UVAQ)
(local university and implementer of 
treatment model)
Center for Youth Integration (CIJ being
its acronym in Spanish) (local NGO and
implementer of treatment model)
Supera (local NGO and implementers 
of treatment model)
Total invested amount: 12,827,153.65 MXN
• Subcontract to C-230 Consultores to develop the Civic Justice 
Homologated Model together with the Mexican government
and the municipal governments of Escobedo and Monterrey: 
$2,790,531
• Subcontract to Observatorio Ciudadano de Seguridad, Justicia y 
Legalidad to develop a training of public servants in PoP and 
mediation in Morelia, Ciudad Juárez, Chihuahua, and 
Guadalajara: $3,826,969.90
• Purchase order to Inteligencia Pública en el Análisis Político to 
support the implementation of the model in Chihuahua and 
consolidate the model in Morelia and Escobedo: $2,017,257
• Purchase order to ProSociedad to support the transfer of the 
model to organizations providing services to infractors in the 
civic justice process in Morelia: $600,411.60
• Grant to UVAQ in Morelia to adapt ProSociedad’s intervention 
model to assist infractors derived from the civic courts: 
$1,116,877.08
• Grant to CIJ in Morelia to implement an addiction treatment 
model to assist infractors derived from the civic courts: 
$701,436.45
• Grant to Supera in Monterrey to adapt ProSociedad’s 
intervention model to assist infractors derived from the civic 
court: $1,773,670.62
* Additionally, the local governments of Morelia and Escobedo co￾invested 14,859,826 MXN and 25,007,475 MXN, respectively, for 
the implementation of the CJHM.
60
Furthermore, the credibility also brought attention and interest from relevant partners, 
such as the private sector in Morelia. Securing co-investment and interest from key players 
favors intervention sustainability, and it also shields intervention continuity from 
government changes, as was also the case in Morelia. 
IV.5.2 SCALING-UP CHALLENGES
Although JPV was remarkably successful in supporting the adoption and scale-up of the ROLE, 
Mente Libre, and civic justice models, rushed expansion, absence of rigorous evaluations, weak 
technical capabilities in implementing organizations, and lack of consideration for differences in 
local circumstances posed challenges for the fidelity and effectiveness of interventions.
Rushed expansion 
JPV’s efforts succeeded in building traction and local demand for the expansion of civic justice 
and CBT models. However, the transfer and expansion processes intensified during the last 2 
years of the JPV activity. Because this left a relatively short time to achieve results, we find that 
the expansion processes were rushed and did not allow sufficient time for polishing 
implementation processes or enabling the consolidation of key intervention components. As a 
result, some models are being consolidated as expansion grows. As two interviewees expressed:
“There is a pressure to scale really quickly. If the first years the [program’s] focus would 
have been on polishing the good things that happened in each municipality before scaling, 
I think that the impact would have been higher. Many of these initiatives require 
maturation time, there are many mistakes that are avoidable if you do not scale up so 
quickly.”
“Suddenly there is a pressure [to scale] from the international agenda. They want many 
numbers and many impacts. If they designed multiannual plans where they had the 
possibility to work very well on one methodology in a few places, and then scale it up, I 
think they would have a bigger impact.”
In the case of ProSociedad, we find that the school context implementation happened with many 
unanswered questions about the theory of change. As one interviewee put it:
“The main design problems were in the causality relationships, there were many causal 
relationships and there were errors in how the lines of causality were drawn…There was 
no clear logic to decide which dynamics to work on during the sessions…Theory was 
used to justify the program, but the program did not have a strong theoretical framework 
and there was confusion between the different program components…It is a very sellable 
program…but a lot of work is needed on the model’s specifications.”
In the case of civic justice, we find that key model components, such as the screening instrument 
and youth referrals to treatment alternatives, were not sufficiently piloted in Morelia and 
61
Escobedo before expansion. Box 6 presents a detailed analysis of the challenges involved in 
scaling up the CJHM. As mentioned by one interviewee:
“The civic justice system is under construction and many procedures are needed. When 
there are no procedures to respond to specific circumstances, we end up responding by 
occurrence…There are protocols, guides, manuals and procedures that still need to be 
developed…There is a lack of coordination between the institutions to serve young 
people, many times youth do not comply with their sanctions.”
Scale-up strategies should prioritize giving organizations sufficient time to refine and consolidate 
all the components of model implementation before scaling them up. This step is essential because 
transferring suboptimal models constitutes a risk to the credibility in the medium term and to 
intervention effectiveness.
Box 6. Key Components of the Civic Justice Homologated Model and 
Implementation Challenges
As the CJHM is rapidly expanding nationally, the challenges of implementing it at scale need to be 
addressed in more detail. One of the key premises of this model is that it helps to deescalate future 
conflicts and to reduce recidivism by offering a means to identify and treat higher-risk offenders. As 
depicted in Figure 3, the CJHM’s preventive approach depends on the effectiveness of the judge’s risk 
assessment, the referral process to adequate treatment services, and the adherence of the offender to 
the treatment services. The case of Morelia provides important lessons about the implementation of 
this model and the challenges involved in securing the correct functioning of its processes. 
First, the civic justice system processes tens of thousands of offenders, with the majority being 
administrative faults such as parking in a restricted area and public noise disturbance, among others. 
Therefore, the capacity of public officials in the civic courts to identify high-risk youth is crucial and
requires awareness about youth risk factors. Furthermore, to determine the required treatments, finer 
assessments tools are needed. The referral process is complex and requires not only strong diagnosis 
skills, but also knowledge of the potential treatment options. As one interviewee put it:
“The probability that an at-risk youth will arrive through the civic justice system is high. Since 
treatment alternatives are expensive, we should focus on those at higher risk. In this sense, risk 
assessment is crucial to prioritize the provision of treatment.”
The case of Morelia shows that these processes are highly complex. Lack of protocols, lack of staff 
training, and limited local capacity to identify the risk factors of offenders passing through the system 
have directly affected the treatment process. For example, the Center of Social Services (CAS being its
Spanish acronym) encountered the challenges of receiving ineligible youth and not having enough group 
participants to initiate therapy groups given the long time taken by the referral process. Municipalities 
where the CJHM has been transferred are facing these same challenges. For example, an interviewed 
public official mentioned that because of the lack of capacity to assess all offenders (more than 30,000
offenders for 1 social worker), judges learn informally along the way how to prioritize treatment 
services. As he put it:
“The main program is the lack of personnel. We just have one social worker. In 2019, over 
30,000 people were presented to the civic court. Judges have learned to prioritize the 
62
56 In particular by giving grants to UVAQ in partnership with ProSociedad to transfer the ROLE model to CAS and CIJ. 
treatment for those who have the smaller probability of incidence. The experience gives you 
light to distinguish which cases are savable.”
Second, the CJHM faces important challenges in ensuring program adherence and in having adequate 
processes to monitor compliance with alternative solutions. Following the example of CAS, this 
experience showed irregular turnout. The lack of social workers’ specialized training results in referrals 
that do not meet offenders’ needs and in the need to rechannel offenders, which significantly hinders 
offenders’ willingness to participate. To address these challenges and through a very intensive technical
assistance process, JPV supported the creation of the Directorate for the Execution of Administrative 
Sanctions, a specialized unit in charge of referral and monitoring processes, as well as worked to adapt 
the YTT for its application within the civic justice system. 
Third, the possibility to refer offenders to treatment services assumes that there is a supply of those 
programs. The case of Morelia showed that this is not always the case. JPV recognized this challenge 
and during 2019 and 2020 launched a strategy to create capacity in local CSOs to offer therapeutic￾component interventions to youth derived from the civic justice system.56 Furthermore, the program 
recognized the need to expand the responsibilities of the Directorate for the Execution of
Administrative Sanctions to congregate other actors of the local system. While these efforts show 
preliminary success, the offer of therapeutic components for youth derived from the CJHM is still highly 
reliable on JPV support, casting doubt on the model’s sustainability. This issue is also found in other 
regions, where personnel from the civic justice system expressed a lack of resources and options to 
send offenders to treatment programs. Moreover, questions remain about whether the available 
treatments respond to youth needs. 
Finally, weak information systems can hinder proper monitoring and tracking and, thus, could affect the 
reduction of recidivism. While Morelia has achieved great progress in developing a registry of offenders, 
it is unclear if the data quality allows for keeping track effectively and to what extent service provision 
organizations report relevant information promptly. These challenges are further exacerbated in 
metropolitan areas such as Escobedo, where there is a need for coordination between different 
municipalities and state-level information systems. 
63
Figure 3. Civic Justice Homologated Model
Source: Based on JPV progress presentation, February 2020.
Absence of rigorous evaluation
Accelerated model expansion and project timeline hindered proper evaluation of the models, 
either by not allowing for process and impact evaluations to be carried out or by failing to create 
spaces to address issues raised by evaluation processes before continuing expansions.
In the case of CEDAT’s Mente Libre, it is remarkable that, through a grant, JPV facilitated an 
impact evaluation of the model. Although the sample size is small, this significant investment 
allowed the organization to get more information on how the program was functioning and 
served to increase the evidence base for its expansion. As one interviewee put it:
“The simple fact of having had the opportunity to do an evaluation with 
electroencephalograms, which is very expensive, and to generate very straightforward 
indicators of the impact of the program, such as that the youth who participated reduced 
their impulsive behaviors by 30% compared to a control group, gave us solid evidence to 
continue working the project in other contexts and to present it in forums.”
Police
( Problem oriented)
Civic Judge
Risk 
assessment
Monitoring 
of high-risk 
offenders
Addiction 
treatment
CBT Employment 
opportunities
Canalization of 
high-risk 
offenders Administrative 
fault
Community 
Community conflict
Multisystemic 
therapy
Thousands of 
offenders
64
In the case of ProSociedad’s ROLE, a process evaluation was performed for its implementation 
in schools. This evaluation was disrupted by a delay in the government’s release of the funding 
for the project’s implementation, which affected the quality of the data collected and reduced the 
evaluation timeline. Nevertheless, the evaluation delivered important findings, such as the ones 
highlighted above about the missing pieces in the program’s theory of change. Also, this evaluation 
finds several fidelity issues in which the model was not delivered as needed by the theory to 
achieve changes, mainly because of lack of implementer capacity, lack of supervision protocol, 
and lack of buy-in from facilitators (teachers in this case). We did not find clear instances in which 
the model addressed these issues. Furthermore, the model is being transferred to different 
contexts (not school settings), which most likely present additional fidelity concerns. No other 
evaluation is being run. 
In the case of the CJHM, evaluations are nonexistent. 
Rigorous evaluation is a key element for the transfer, adaptation, and replication processes, and 
it is imperative that the following USAID activities in Mexico prioritize evaluating the 
interventions before continuing their expansion. Process evaluations are needed to ensure 
program fidelity, and impact evaluations are needed to confirm whether the program’s intended 
mechanisms for change actually work.
• For the CBT-based models, ROLE and Mente Libre, although there is a strong evidence
base of effectiveness, further rigorous evaluation is needed to assess which approaches 
are more appropriate and effective in the local contexts. As noted in the USAID J-PAL–
commissioned report, there are important pending research questions about whether 
these models with complementary interventions, such as cash support or more integral
treatment approaches, could be more effective. Also, there is a significant opportunity to 
determine the most cost-effective length of treatment and how to adapt the programs for 
a higher-risk population. 
• For the CJHM, the lack of evidence is worrisome. The evidence that has supported the 
model’s alleged impact is purely anecdotal or based on perception. Yet, it has expanded 
quickly across the country. 
The future expansion of the model to different municipalities offers a timely opportunity 
to rigorously evaluate the CJHM. First, process evaluations are imperative in all 
municipalities to make sure that the key model components are properly implemented. 
Second, impact evaluations of the different components are desirable to understand 
whether and how the mechanisms are working. To this end, the following are needed:
a. Define what we want to learn, what are the relevant questions, what outcomes 
are expected, and how can they be measured. For example, the outcomes of 
interest could range from changes in citizen perception of safety by reducing 
impunity for minor offenses to changes in recidivism of offenders who go through 
the system and receive treatment alternatives to changes in citizen trust in the 
police due to community outreach efforts. 
65
b. Then, an impact evaluation design with a control group should be determined 
considering all the political, administrative, and ethical constraints. For example, 
an experimental phase-in design would guarantee that all units of analysis would 
eventually receive the “treatment.”57
Although rigorous evaluations are not easy to perform because they are time-consuming 
processes and require specialized technical knowledge, it is imperative that models are rigorously 
evaluated at a small scale before expanding them. Otherwise, there is a risk of transferring models 
that have not been tested and that may not have the desired impact.
Weak technical capacities in organizations that adopt the models 
When replicating and expanding promising interventions, developing implementation capacity in 
organizations that adopt the models is crucial to ensure fidelity. Transfer and implementation of 
any policy need to account for the specialized training of operators. 
In the case of CBT, CEDAT provides intensive training and certification to program operators. 
For CEDAT, a priority in the transfer process is the specialized training of facilitators. The transfer 
model includes 100 hours face to face and practical and theoretical training, and approximately 
30% of the trainees earn certification. Unlike CEDAT, the training for program operators who
are adopting the ROLE model is unclear. While the manuals consider training, the parameters 
are quite broad, casting doubt on the extent to which facilitators are sufficiently trained. We 
consider that the key difference is because of CEDAT’s extensive experience working directly 
with youth and with training operators, whereas ProSociedad’s ROLE is a second-floor
organization. Therefore, particularly in cases where there is less experience in implementing, 
process evaluations are of the highest priority. CEDAT’s good training practice is consistent with 
the academic CBT literature, which indicates that training of treatment providers and their ability 
to connect with youth are key to guarantee effectiveness.
58
For interventions such as the CJHM, the lack of adequate training of the staff in charge of 
channeling offenders can affect the model's functioning and undermine its effectiveness. We find 
that the model is being transferred without ensuring proper staff training in the key intervention 
components, as highlighted in Box 6. A factor that could have hindered capacity-building at the 
operator level is that JPV favored a top-down approach during the transfer processes. As 
mentioned by one interviewee:
“JPV worked very top-down. First, we had to talk to the governor, the prosecutor, the 
commissioner, and the big players. That is the most traditional way of doing public policy, 
but we also find a lot of value in going from the bottom up…The governor can say yes 
but the social worker can say no, sometimes the line actors are left aside but that is not 
always the most effective when it comes to practice.”
57 For more information on how to design impact evaluations, see https://www.povertyactionlab.org/sites/default/files/research￾resources/Introduction-to-Evaluations_0.pdf 58 See “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018.
66
Weak adaptation to local contexts 
A crucial element in the adaptation and replication of models to other contexts is to account for
how the new local circumstances affect the underlying model theories. Because no two contexts 
are identical, scaling programs in different places is not as simple as transferring an "out-of-the￾box" solution. In this sense, a program should only be transferred when the new context allows 
for adoption of the essential causal mechanisms of the intervention’s theory of change. For the 
selected interventions, we find that oftentimes the scale-ups were carried out in widely different 
contexts without careful adaptation. As a result, the selected interventions may not be ideal to 
address local risk factors and dynamics of crime and violence.
While JPV succeeded in facilitating knowledge transfer and leveraging the effort invested in 
systematizing the ROLE and Mente Libre models by transferring them to treat offenders referred 
by the civic courts, questions remain about whether these models were properly adapted. Even 
though CBT interventions are relatively easier to systematize (see Table 10) given that they are 
focalized and have a determined curriculum, careful consideration must be given to underlying 
risk factors of different target populations. For instance, youth at school might have different risk 
factors and needs than the target population of the civic justice system, ranging from an out-of￾school, employed 25-year-old sent to a CBT treatment to a 40-year-old given the same treatment
option. Particularly for the case of ProSociedad, it is important to note that the pilot evaluation 
was carried out for the model’s implementation in schools, but it is unclear how this model 
transfers to implementation with high-risk youth identified by civic courts. Moreover, failures in 
the referral process may affect adherence. As mentioned by one interviewee:
“There is no sensitization process by the courts. Offenders arrive to the programs with 
reservations regarding their participation and with the idea that the measure is a 
punishment. This hinders the intervention process, at least during the first sessions.”
For the CJHM, in the absence of a rigorous evaluation, many doubts remain about in which 
contexts it is optimal to adopt the model. The piloting of the model in Morelia and Escobedo was 
carried out in very particular contexts, with specific political dynamics and accompanied by 
substantive reforms in police operations. In this sense, when thinking about expansion, it is key 
to take into account how the underlying theory of the model is affected by differences in crime 
dynamics and institutional capabilities. For example, police officers working in high-violence cities 
may present more hesitation to adopt mediation strategies. Implementing a PoP approach might 
be challenging in places where commanders do not favor cultural change or in places where the 
local police are unable to perform investigative work. Furthermore, the lack of operating rules 
and procedures in some localities generates a disconnect between the objectives pursued by the 
police and the judges. As explained by different interviewees:
"In Morelia we found that the police stopped detaining people because they considered 
the civic justice to be a revolving door model and they did not want to arrest anyone if 
the judges were going to let offenders go."
"In Juárez it has been very difficult to train police officers because they know that 
mediation does not help them when they have to focus on avoiding being killed."
67
"In general, daily police work does not allow the civic justice and PoP model to operate 
because the work of the police continues to depend on public prosecutors and indicators 
such as the number of arrests."
“We have many problems with the Secretary of Public Security because the judges already 
operate the model, but the police have not been willing to implement it. The approval of 
a regulation would allow us to have teeth for the police to cooperate with the model. 
Sometimes we have to talk to superiors so that they order middle-ranking officials to 
follow the implementation of the Civic Justice model.”
While the CJHM could be a relevant and effective alternative in different contexts across Mexico, 
the model should not be seen as a silver bullet for regions with great differences in crime dynamics 
and institutional capabilities.
68
STAGE 2—DID THE PROGRAM CONTRIBUTE TO A REDUCTION OF AT-RISK 
YOUTH LIKELIHOOD PARTICIPATING IN CRIME AND VIOLENCE?
3. What evidence is there that JPV has reduced the likelihood of at-risk youth in 
targeted communities participating in crime and violence?
a. What evidence is there that at-risk youth reached with JPV programming 
will engage in alternative, positive pathways and steer away from 
involvement in crime and violence, including with TCOs?
b. Is JPV’s coverage or reach of at-risk youth in targeted communities (i.e., 
number of at-risk youth reached) sufficient to initiate a detectable 
reduction in youth involvement in crime and violence? Why or why not?
c. To what extent have gender-based roles, including but not exclusive to the 
mothers of at-risk youth, affected the ability of the interventions to achieve 
their stated results (effectiveness)?
Box 7: Question 3 Summary
To answer this question, we analyze whether JPV’s coverage or reach of at-risk youth in targeted 
communities was sufficient to initiate a detectable reduction in youth involvement in crime and violence. 
To this end, we gathered, to the extent possible, data from beneficiary databases from supported 
interventions. The results presented herein are preliminary. At a later stage, we will perform focus 
groups with the beneficiaries of direct interventions to understand which components might have 
worked or not and why. We will also collect data of at-risk youth in the targeted municipalities through 
a representative sample of communities with high homicide rates to measure individual-level exposure 
and attitudes toward violence.
Preliminary conclusion and next steps: Our field research and evidence review suggest that crime 
in Mexico is a highly contextual, socio-culturally driven behavioral phenomenon. Therefore, a successful 
intervention also needs to take into account the broader context of violence to which program
participants continue to be exposed during and after the program. However, our amassed secondary 
information shows that JPV-supported interventions usually only targeted 1 out of the 10 most violent 
neighborhoods within the municipalities of interest, and often targeting only a limited number of youth
in these neighborhoods. Our preliminary results suggest that, while the intervention may have had a 
positive effect on individual participants in those municipalities, a wider coverage, both in terms of 
services offered and youth reached, would be necessary in order to obtain community-level effects in 
crime and violence reduction.
In terms of services offered, we find that the designs of existing interventions do not sufficiently tackle
youth risk factors and that gender roles are barely addressed. The programmatic offer currently covers 
impulsive behavior, anger management, and employability skills, and to a lesser extent substance 
consumption. However, data from the Youth-Targeting Tool (YTT) confirm that a wider range of 
interventions is needed to address youth risk factors, particularly to promote protective factors. 
Interventions that are able to promote protective factors and interrupt the community and family 
dynamics that produce antisocial behaviors remain urgent, as does expanding the offer of substance 
abuse treatment.
69
IV.6 COVERAGE AND TARGETING
In terms of the number of at-risk youth reached by the JPV-supported interventions, we can only 
know the number of youth who have received at least one intervention by municipality, but we 
cannot speak to the effectiveness of these interventions (up to this point). Table 12 shows the 
number of beneficiaries by region.59 The number of beneficiaries is much higher in Ciudad Juárez
and Monterrey than in the rest of the regions. But if we want to understand whether the level of 
coverage is enough to initiate a detectable reduction in youth involvement in crime and violence, 
then we need a different type of analysis. Following the evidence that crime is concentrated in 
certain people, places, and behaviors, we need to understand whether benefited youth come 
from high-risk places. We were able to perform this analysis by linking data of the beneficiaries’ 
neighborhoods of residence, available from intervention databases, with neighborhood-level 
homicide data that we requested from local authorities.
60 For this analysis, we worked with the 
individual databases of JPV-supported interventions. One important caveat is that the YTT does 
not have beneficiaries’ neighborhoods of residence, and intervention databases do not have 
complete information of the YTT, which prohibits crossing the information of risk factors by 
neighborhood or even by municipality.61 Also, the data are incomplete, and we have several youth 
for whom neighborhood of residence is not specified. 
Table 12. Beneficiaries by Gender and Region
Municipality Male Female Total
Chihuahua 76 32 108
Monterrey 416 138 554
Guadalajara 97 9 106
Ciudad Juárez 548 391 939
Tijuana 171 145 316
Morelia N/A N/A
Other 150 214 216
Total 1,458 929 2,023
In Annex VIII, we present this targeting analysis, for which we rely on the data provided by local 
authorities. It is important for the reader to keep in mind that these numbers do not coincide 
with those reported by Mexico’s National Institute for Statistics and Geography (INEGI being its 
acronym in Spanish) or by the Mexico’s Secretary General of National Public Security (SESNSP
being its acronym in Spanish). In Annex VIII, we present a comparative table showing the totals 
between these numbers, and then we show the top 10 neighborhoods in terms of homicides and
the most common beneficiary neighborhoods registered in the intervention databases for Ciudad 
Juárez, Chihuahua, Tijuana, Escobedo, and Guadalupe. Unfortunately, we do not have data of 
59 We got these numbers from intervention databases. See footnote 58. 60 We requested homicide data as follows: for Monterrey, Guadalupe, and Escobedo, we requested the data from the Nuevo 
Leon State Attorney General; for Ciudad Juárez and Chihuahua, we requested the data from the Chihuahua State Attorney 
General; and for Tijuana, we requested data from the Municipal Security Secretariat. 61 We tried to merge the YTT with the intervention beneficiary databases, but unfortunately, the individual identifiers do not 
coincide. It is also important to note that the total numbers of beneficiaries in both sets of data do not coincide. There are more 
beneficiaries in the individual databases than there are in the YTT, but still, for some interventions we do not have access to their 
database. 
70
homicides at the neighborhood level for Morelia or any municipality of the metropolitan area of 
Guadalajara, and we do not present analysis for Monterrey because we have very few beneficiary
observations with neighborhood of origin. The analysis shows that JPV-supported interventions 
work with only 1 of the top 10 neighborhoods in terms of homicides in Ciudad Juárez, Chihuahua, 
Tijuana, and Escobedo and work with 2 in Guadalupe. Moreover, our field research and evidence 
review suggest that crime in Mexico is a highly contextual, socio-culturally driven behavioral 
phenomenon.
62 Therefore, a successful intervention also needs to take into account the broader 
context of violence to which program participants continue to be exposed during and after the 
program. Our preliminary results suggest that, while the intervention may have had a positive 
effect on individual participants in those municipalities, a wider coverage, both in terms of services 
offered and youth reached, would be necessary in order to obtain community-level effects in 
crime and violence reduction.
In terms of services offered, we analyzed the data available from the YTT and find that the designs
of existing interventions do not sufficiently address youth risk factors, also confirming the 
interviews findings highlighted in the previous section. Table 13 shows the risk factors, ordered 
from the highest occurrence to the least for male beneficiaries, and Table 14 shows those for 
female beneficiaries. To build these tables, we summed the scores of each risk factor for all 
beneficiaries available in the database.63 The YTT scores show how acute each risk factor is for 
each beneficiary as follows: 0 is low, 1 is moderate, 2 is high, and 3 is critical. 
These tables confirm that that the top five risk factors are not necessarily covered by the 
programmatic offer, and moreover, they show important differences between genders. For both 
genders, within the top five risks, we find insecurity in the community and lack of parental figure. 
For females, among the top five risks factors, we find violence at home, lack of social support, 
and anger management, while for males we find difficulties in the school context, substance 
consumption, and impulsive behavior. The programmatic offer currently covers impulsive 
behavior, anger management, and, to a lesser extent, substance consumption, but data from the 
YTT confirm that a wider range of interventions is needed to address youth risk factors,
particularly to promote protective factors. In this sense, interventions that are able to interrupt 
the community and family dynamics that produce antisocial behaviors are imperative, as is 
expanding the offer of substance abuse treatments. 
62 See Bandura and Walters, 1977 and “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 
2018.
63 We use data available for the YTT version 2, as this has significantly more beneficiaries. We did not sum version 1 and version 
2 because not all variables are comparable. 
71
Table 13. Male Beneficiaries’ Risk Factors 
(YTT V.2)
Risk factor
Total 
general
Insecurity in the community 664
Difficulties due to substance 
consumption 580
School context 550
Lack of parental figure 498
Impulsive behavior 492
Peer crime involvement 486
Early start in violence 485
Breeding style 464
Anger management 464
Perceived social support 461
Exposure to violence at home 437
Attitude toward violence 421
History of childhood abuse 343
Physical violence 325
Conflict with the law 313
Empathy/regrets 300
History of nonviolent crime 293
Goal fulfillment 252
Criminality of parents or caregivers 228
Lack of attention / hyperactivity 228
Table 14. Female Beneficiaries’ Risk Factors 
(YTT V.2)
Risk factor
Total 
general
Insecurity in the community 400
Lack of parental figure 336
Exposure to violence at home 304
Anger management 291
Perceived social support 277
Breeding style 270
Impulsive behavior 260
History of childhood abuse 234
School context 230
Attitude toward violence 215
Difficulties due to substance 
consumption 215
Early start in violence 201
Peer crime involvement 194
Goal fulfillment 148
Physical violence 140
Criminality of parents or caregivers 131
Lack of attention / hyperactivity 131
Empathy/regrets 92
History of nonviolent crime 77
Conflict with the law 66
The YTT data also show that gender roles affect intervention effectiveness, as misogynistic and 
established masculinity beliefs and values reinforce risk factors, such as violence normalization, 
particularly at home. However, gender roles were barely addressed by supported interventions. 
The program’s gender strategy was managed in parallel to the at-risk youth transversal strategy. 
To our knowledge, the only program that incorporates a more comprehensive gender 
perspective is Supera (with masculinity workshops and work with mothers and women 
neighbors), and CASA has masculinity workshops. Nevertheless, in several interviews, this topic 
came up as pending priority area. As two interviewees put it:
“Once we organized a mixed soccer league, in which the team had to be conformed by 
three men and three women. We had to work a month and a half for youth to understand 
why.”
“Another factor is associated with gender. We refer to the violence that some young 
women suffer from. Here we have several women that are used as shields; that is, they 
are the ones who carry the drugs and weapons to transport them from one place to 
another. Young women victims of abuse. As they grow up, they become part of the lines 
of organized crime.”
While all of these data give us some insights, we need to understand youth perceptions, behaviors, 
and attitudes toward violence. At a later stage we will perform focus groups with direct 
beneficiaries of JPV-supported interventions to explore which components may have worked or 
not and why in reducing the likelihood of youth being involved in crime and violence. We will 
72
also collect data of at-risk youth in the targeted municipalities through a representative sample 
of communities with high homicide rates to measure individual-level exposure and attitudes 
toward violence. This survey will be informative of the potential mechanisms through which the 
intervention may have affected the outcomes of interest. 
IV.7 JPV-SUPPORTED INTERVENTIONS RESULTS IN CRIME AND VIOLENCE 
REDUCTION IN TARGETED LOCALITIES
4. What evidence is there that the USAID/Mexico crime-and-violence-prevention 
programming through the JPV activity has contributed to achieving the goal of 
reducing crime and violence within the targeted JPV communities?
c. Where and what are examples of the JPV activity contributing to reduced 
levels of crime and violence in the targeted municipalities?
d. To what extent have citizen perceptions of security increased in targeted 
municipalities?
Box 8: Question 4 Summary
To answer this question, we perform a descriptive statistical analysis of existing secondary data on 
crime and homicide rates, as well on citizen perceptions of security. 
Preliminary conclusions and next steps: Our initial analysis finds that homicide rates present 
positive trends for all municipalities of interest since 2015. Insecurity perceptions keep rising in all 
municipalities, except for Guadalupe and Escobedo (although data for these municipalities are only 
available since June 2018). However, we remain very cautious and refrain from drawing any causal 
connections from this analysis. The increased levels of violence in Mexico are in large part attributable 
to organized crime. Consequently, insecurity perceptions are also driven by violence related to drug 
cartels and to other highly visible acts of violence that affect perceptions. Hence, a decrease in homicide 
rates and reducing insecurity perceptions, may require different strategies than the one envisioned for
JPV and that entail longer-term institutional reform and more traditional law enforcement action.
The National Survey of Urban Security (ENSU being its acronym in Spanish) provides other available 
indicators related to antisocial behaviors that may be better suited to explore program contribution to 
reducing the likelihood of youth engaging in crime and violence. Unfortunately, the lack of clear 
counterfactual information, along with limited socio-economic data, hindered our ability to isolate the 
effect of the program on youth population. From our field explorations we generate a plausible 
hypothesis that JPV could have impacted individual youth perceptions, attitudes, and exposure to 
violence. In order to address the challenges with the existing data, we have designed a unique and 
tailored survey of youth that will measure youth exposure, attitudes, and behaviors toward violence. 
Our stratified sampling procedure will include youth in the targeted municipalities as well as youth from 
similar municipalities which remained outside the realm of JPV, to form a comparison group.
We perform a descriptive statistical analysis of existing secondary data on homicide and other 
crime rates, as well as on citizen perceptions of security, to understand the context in which this 
program was implemented and the observable trends in treated municipalities and the rest of the 
country. However, we remain very cautious and refrain from drawing any causal connections 
from this analysis. Increased levels of violence in Mexico are in large part attributable to organized 
crime. Estimations from different sources agree that over one-third of homicides are related to 
73
this source of violence.
64 Consequently, insecurity perceptions are also driven by violence related 
to drug cartels and to other highly visible acts of violence that affect perceptions. Addressing this 
type of violence requires different strategies than the one pursued by JPV. A decrease in 
homicides rates and in general insecurity perceptions would require different strategies that entail 
longer-term institutional reform and more traditional law enforcement action.
65 
In terms of homicides, we built a database that includes the monthly occurrence at the 
municipality level from 1990 to 2018. We created this database with data from INEGI, which uses 
vital statistics based on death certificates; the data tend to be more precise than those reported 
by SESNP.66 Nevertheless, in our database, we also include homicide statistics from SESNP from 
2015 to 2020 to be able to also observe patterns in 2019 and 2020 (INEGI data, unfortunately, 
are published with over a year of delay). Our database also includes statistics on other crimes 
from SESNP (car robbery with and without violence, extortion, and kidnapping), and we also 
include other sociodemographic characteristics at the municipality level, such as age distribution, 
marginalization index, employment, and education from INEGI and the National Population 
Council (CONAPO being its acronym in Spanish). Finally, we have built an additional database, 
drawing on data from ENSU quarterly from January 2016 to March 2020. We have performed an 
initial descriptive analysis, for which graphs are available in Annex IX. We find that homicide 
rates present positive trends for all municipalities of interest since 2015. Insecurity perceptions 
keep rising in all municipalities, except for Guadalupe and Escobedo (although data for these 
municipalities are only available since June 2018). 
The ENSU provides other available indicators related to antisocial behaviors that may be better
suited to explore program contribution to reducing the likelihood of youth engaging in crime and 
violence, for instance, perception data on alcohol and drug consumption and sale in public spaces, 
the existence of street gangs,67 and the percentage of people who report engaging in conflict. 
These data show mixed trends. For alcohol consumption in public spaces, trends are all negative 
(including the national average) for all municipalities of interest except Tijuana. Conversely, drug 
consumption and sale perception tendencies are positive for Tijuana, Morelia, Monterrey, and
Ciudad Juárez, and are negative for Chihuahua, Escobedo, Guadalupe, Tonalá, and Zapopan (the 
national average tendency is slightly positive). The perceptions of the existence of street gangs 
have barely changed during the last 4 years, and the rates of people who report engaging in 
conflict are negative for all municipalities (the national average is also negative).68
It is important to keep in mind that that the intervention was highly heterogeneous, as 
implemented prevention strategies varied by municipality. Some had a greater focus on civic 
justice, while others focused on reintegration and territorial strategies. Therefore, a more 
nuanced analysis of mechanisms would be needed for each case. Furthermore, these data present 
three additional challenges. First, the sample is only representative of the population who are 18 
years and older. Second, the data are based purely on perceptions, which may have respondent 
64 See Calderón et al., 2018. 65 See “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018. 66 For a more nuanced analysis of the differences in crime reports and potential underreporting, see Diego Valle - Jones blog: 
https://elcri.men/en/underreporting-crime/ 67 Pandillerismo in Spanish; there is no exact translation for this term. 68 For Tonalá, Zapopan, Escobedo, and Guadalupe, data are only available since June 2018. 
74
bias. Third, the data measure perceptions on the existence of drug and alcohol consumption and 
street gangs, but do not directly measure youth participation in these activities. Given all of this 
and the lack of a counterfactual, the effect the program may have had directly on youth cannot 
be disentangled conclusively. The preliminary statistics we present aim only to provide the reader 
with a descriptive overview of the trends in the municipalities of interest.
From our field explorations we generate a plausible hypothesis that JPV could have impacted 
individual youth perceptions, attitudes, and exposure to violence. In order to address the 
aforementioned challenges to the extent possible, we will conduct a youth survey that will 
measure youth exposure, attitudes, and behaviors toward violence. This sample will be 
representative of youth in the municipalities of interest and will also include a set of similar 
municipalities to descriptively compare outcomes with a more representative view. We will build 
this sample with data gathered from INEGI and CONAPO. Finally, we will address potential 
respondent bias by embedding a vignette experiment into the survey to measure the effect of 
hypothetical exposure to the intervention on subjects’ attitudes toward criminality. As explained 
in the Methodology sections, vignette studies use short descriptions of hypothetical scenarios 
(vignettes) that are usually presented to respondents within surveys in order to elicit how their 
judgments about such scenarios affect outcomes of interest, often revealing their perceptions, 
values, or social norms. This survey will help identify potential mechanisms through which the 
intervention may have affected the outcomes of interest. 
75
V. RECOMMENDATIONS
The purpose of this performance evaluation is to gauge the effectiveness of USAID’s JPV program 
in promoting reductions in levels of crime and violence and gather recommendations for 
improved future programming. In particular, this evaluation provides recommendations on more 
impactful, proven, and sustained approaches for future crime-and-violence-prevention 
programs. Based on JPV’s learnings, we provide the following recommendations to inform future 
policies.
1. Define narrower systems for crime and violence prevention.
The local systems framework provides a practical perspective to engage and empower local
actors in collaborations to achieve specific and sustained development results. One key 
feature to make the local systems framework successful is to build such systems around
definite outcomes. We recommend that the result structures respond to the program’s 
immediate outcome rather than the program’s long-term development goal. Particularly in 
issues as complex as crime and violence prevention and in settings where the drivers are too
diverse, systems must address narrower results. Otherwise, they will run the risk of being 
too broad, making responsibilities and contributions confusing, thereby collaborations weak. 
Furthermore, the result should be commensurate with the program’s reach. To achieve this,
the local systems can be defined around the goals of reducing youth participation in violence, 
construction of a culture of legality, or reducing gender violence. We deem that these 
narrower goals would allow for more effective interaction with the systems. Additionally, 
concentrated efforts may also decrease the number of conduit organizations increasing 
bureaucratic efficiency as well as reducing costs. A few corollary recommendations also 
emerge.
a. Facilitate communication channels between specialized systems that 
contribute to the same long-term development goal. Different systems 
might be contributing to the same development goal. The local systems approach 
should consider facilitating communication channels between systems. This 
strategy could leverage the potential overlap of a few actors between systems. For 
instance, the executive branch of a municipal government might take part in all 
three of the systems highlighted above as examples.
b. Work with diverse leadership. The strategy to engage with the system must 
respond to natural leadership and should leverage comparative advantages. More 
specialized systems also facilitate the recognition of leadership, which allows for a 
finer engagement. 
c. Develop local agendas around specific strategies with a clear technical 
offer, defined target populations and indicators, and clear responsibilities and goals.
d. Generate networks of specialized knowledge. These networks should 
respond to the accumulation of knowledge about techniques, models, and tools 
to work with the targeted goals, for instance, for reintegration strategies or 
domestic violence. 
e. Program evaluation indicators should respond to narrower goals. The project 
monitoring, evaluation, and learning plan for activities that involve local systems 
76
should respond to the narrower goals (the same as those around which the 
systems will be organized) rather than to the long-term development goal. To this 
end, it is worth working with the theory of change and thinking backward about 
the steps needed for a required change. In this mapping, the links between the 
program activities and expected outcomes should become clear.
2. Focus efforts on generating a wider range of models for at-risk youth by giving 
clear tools to local actors for catalyzing new models and scaling up.
The aim should be to enable the environment for specific and promising innovations to be 
able to thrive, to generate knowledge, and to identify new effective solutions with the 
potential for scaling. Unfortunately, the evidence base on what works to reduce violence 
among youth is relatively narrow. Governments, CSOs, researchers, and the private
sector are addressing this issue by conducting a variety of policy experimentations. The 
objective of aid programming should be on developing specific subject-matter related 
technical expertise, implementation, and M&E capacities. In particular, investment should
achieve the following.
a. Develop intervention models adapted to the local context on 
interventions for which there is a strong evidence base, in particular 
multidimensional therapy models, addiction treatment, and psychiatric 
approaches. Given the risk factors presented for youth, the rapid development 
of these models is imperative. 
b. Foster innovation based on the concentration, coordination, and 
proactivity principles. When the existing evidence base of effective 
interventions is not enough, the approach should be to leverage local innovations 
and to provide the necessary resources to solidify models following these key 
principles. 
3. Adapt, pilot, and then scale.
Evidence-based policy design consists of a constant process of adaptation and innovation, 
testing, and scaling up. As explained in the USAID J-PAL–commissioned report, “Programs 
that are effective at improving outcomes in one context may help in other places where 
the key problems and underlying reasons for the problems are similar. But since no two 
contexts are identical, scaling programs in a different place is not as simple as transferring 
an ‘out of the box’ solution. To scale appropriately and effectively, policymakers and 
researchers must account for local conditions and implementation capacity, and then
adapt programs to that new context accordingly.”69 It is imperative that international 
cooperation initiatives take careful consideration of the need to adapt programs. Even 
within a country, local contexts differ vastly. The risk of scaling up without the adaptation 
piece relies on the effectiveness of the intervention and, as a consequence, may harm 
credibility and prevent further adoptions. In particular, when considering scaling up, it is 
necessary to:
69 For more information, see https://www.povertyactionlab.org/evidence-to-policy/adapting-and-scaling-program
77
a. Allow enough time to refine implementation. Organizations should be 
provided sufficient time to refine and consolidate all of the model components 
before scaling up a strategy. This will ease transfers and secure fidelity by 
establishing protocols to monitor whether the implementation follows the desired 
theory of change. 
b. Foster comprehensive processes of replication that consider differences 
in context. When adapting a program, careful consideration should be given to 
the local context and to whether the theory of change applies to the new 
circumstance. It is important to verify if the conditions are conducive for the 
implementation to be carried out as desired. If the local conditions allow for 
adoption of the essential components of the theory of change, then the 
nonessential components should be adapted to the local context. If the essential 
components cannot be held under the new circumstance, then other policy 
alternatives should be explored.
70
c. Invest in a few strong pilots, which might bring a higher payoff than 
scaling up suboptimal models. International cooperation can play a 
groundbreaking role by catalyzing innovations at the local level and shifting 
incentives to attract the attention of other actors including the government and 
private sector, which have the power of investing at scale. In this sense, the 
investments of programs such as JPV may bring higher payoffs by nurturing 
promising interventions, enabling the conditions for them to thrive and generating 
rigorous evidence about their effectiveness. These strong and evidence-based 
examples can then become catalysts of further innovations in the medium and long 
terms. The approach should be to rigorously evaluate pilots and generate data 
driven evidence of success at a small scale before expanding models. 
d. Evaluate. Rigorous evaluation is a key element for the transfer, adaptation, and 
replication processes, and it is imperative that the following USAID activities in 
Mexico prioritize evaluating the interventions before continuing their expansion. 
Process evaluations are needed to ensure program fidelity, and impact evaluations 
are needed to confirm whether the program’s intended mechanisms for change 
work as expected.
4. Engage with government agencies to create specialized capacity for the design 
and implementation of specific services aimed at reducing recidivism. 
Increasing the capacity of CSOs to work with at-risk youth is paramount, especially given 
that the civil society sector has more flexibility to try out innovative approaches. 
However, there is also a need to generate specialized capacity within governments to 
support youth to build alternative pathways to crime and violence is of the greatest 
priority. In particular, it is imperative to achieve the following:
a. Generate capacity within the police and judicial systems. Police 
departments and the judicial system need to shift away from a public safety and 
70 See “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018.
78
punitive approach to a public health perspective to prevent youth crime and 
violence. This approach should be based on an understanding of youth protective 
and risk factors. Law enforcement and judicial prosecutors need to work jointly 
with service providers to facilitate the long-term recovery of offenders. Working 
with these institutions to further this approach is a necessary condition for 
prevention policies to thrive. There is an enormous area of opportunity to work 
with the Mexican government at all levels to translate the National Juvenile System 
Law approved in 2016 into real policy alternatives for youth. There are 
opportunities to augment the judicial system’s human capital and technology and 
to develop clear protocols and channels to work with youth. The aims should be 
to avoid the “revolving door” phenomenon for minor offenses by creating a system 
that supports youth while keeping them accountable. The judiciary needs to 
ensure that prosecuted cases send youth to the services they need (more on this 
below).
71
b. Generate capacity within governments for the provision of social 
services to at-risk youth. Related to the above, combining post-release or 
alternatives to incarceration with comprehensive social services that are aimed at 
remedying the behavior leading to the original arrest and that promote safety, 
strength, and permanence in the community, schools, home, or employment while
keeping youth accountable72 is the most promising approach to reduce recidivism 
sustainably. These services should coordinate health and mental health services, 
education and employment support, and family counsel. There remain ample 
opportunities to engage with public institutions to generate this specific offer.
c. Facilitate the development of information systems. To better diagnose and 
target policies, governments need to significantly increase their technological 
capacities to gather and share data, and international cooperation is well 
positioned to support this goal. Data should include, among others, georeferenced
incidents, settings and circumstances of incidents, deaths due to violence, and 
nonfatal violence-related injuries. Information systems should incorporate data 
from different departments including police, justice, labor, education, and health.73
Moreover, it is paramount that data systems are comparable between 
municipalities and that data are shared in bigger geographical areas. This is 
particularly relevant for metropolitan areas. 
d. Facilitate youth case follow-up. There is enormous room to generate 
information systems that facilitate youth follow-up and coordination between 
agencies and CSOs. While the YTT is a good start, we recommend making the 
platform more flexible. That is, instead of having predetermined variables, such a 
platform could allow organizations to define their own variables, on top of 
choosing from an existing pool, to define the monitoring protocol that makes 
more sense. By making this platform more flexible, institutions will also be able to 
further exploit existing information.
71 See “What Works to Prevent Violence Among Youth?”, USAID J-PAL–commissioned report, 2018. 72 See Cook et al., 2015. 73 World Health Organization, 2015.
79
REFERENCES
Abt, Thomas P. 2017. “Towards a framework for preventing community violence among youth.” 
Psychology, Health & Medicine 22 (1): 266-285. doi:10.1080/13548506.2016.1257815
Bandura, Albert, and Richard H. Walters. Social learning theory. Vol. 1. Englewood Cliffs, NJ: Prentice-hall, 
1977.
Blattman, Christopher, Julian C. Jamison, and Margaret Sheridan. 2017. "Reducing Crime and Violence: 
Experimental Evidence from Cognitive Behavioral Therapy in Liberia." American Economic Review 107 
(4): 1165–1206.
Calderón, Laura, Octavio Rodríguez and David A. Shirk. 2018. “Drug Violence in Mexico: Data and 
Analysis Through 2017” Justice in México Project (april). 
https://justiceinmexico.org/wp-content/uploads/2018/04/180411_DrugViolenceinMexico-12mb.pdf
Cook et al., 2015, An Experimental Evaluation of a Comprehensive Employment-Oriented Prisoner Re-entry 
Program, Journal of Quantitative Criminology. 
Easton, Caroline J., Dolores Mandel, Karen Hunkele, Charla Nich, Bruce Rounsaville, and Kathleen Carroll. 
2007. “A cognitive behavioral therapy for alcohol-dependent domestic violence offenders: an integrated 
substance abuse-domestic violence treatment approach (SADV).” The American Journal on Addictions 16 (1): 
24-31. 
doi:10.1080/10550490601077809
Friedman, Alfred S., Arlene Terras, and Kimberly Glassman. 2008. “Multimodel Substance Use Intervention 
Program for Male Delinquents.” Journal of Child & Adolescent Substance Abuse 11 (4): 43-65. 
doi: 10.1300/J029v11n04_03
Heller, Sara B., Anuj K. Shah, Jonathan Guryan, Jens Ludwig, Sendhil Mullainathan, and Harold A. Pollack. 
2017. "Thinking, Fast and Slow? Some Field Experiments to Reduce Crime and Dropout in Chicago." The 
Quarterly Journal of Economics 132 (1): 1–54.
Office of Research, “Federal Funding (Grants vs. Cooperative Agreements vs. Contracts)”, UC Santa Cruz. 
https://officeofresearch.ucsc.edu/osp/prep/fed-funding-mechanisms.html
USAID, “Evidence Base for Collaborating, Learning and Adapting. A summary of the literature review”, 
First Published March 2020. 
https://usaidlearninglab.org/sites/default/files/resource/files/eb4cla_summary_literaturereview_20171020
_final.pdf
USAID, “Executive Summary: Estimation of the costs of crime and violence for the private sector in 
Mexico and the importance of its participation in violence prevention as a strategy for its long-term 
business model”. First Published June 2019.
USAID, “Grant and Contract Process”, Last updated January, 2020.
https://www.usaid.gov/work-usaid/get-grant-or-contract/grant-and-contract-process
80
USAID J-PAL–commissioned report, “What Works to Prevent Violence Among Youth? A White Paper 
on Youth Violence, Crime Prevention, and the Mexican Context.” First Published September 2018.
http://www.cca.org.mx/forocdmx/pdfs/White%20Paper%20JPAL_Ingles_Final.pdf
USAID, “Local Systems: A Framework for Supporting Sustained Development”, First Published April 2014. 
https://www.usaid.gov/sites/default/files/documents/1870/LocalSystemsFramework.pdf
USAID, “Performance Evaluation Designs”, First Published September 2019. 
https://www.usaid.gov/project-starter/program-cycle/project-monitor-evaluation-plan/monitor￾evaluation-plan-evaluation-component/performance-evaluation-designs
USAID, “The 5Rs Framework in the Program Cycle”. First Published October 2016.
https://usaidlearninglab.org/sites/default/files/resource/files/5rs_techncial_note_ver_2_1_final.pdf
USAID Learning Lab, “Introduction to Collaborating, Learning and Adapting (CLA) in the Program Cycle”, 
2019.
https://usaidlearninglab.org/course-introduction-cla-program￾cycle#:~:text=Collaborating%2C%20Learning%2C%20and%20Adapting%20(,their%20Journey%20to%20S
elf%2DReliance.
Vaughn, Michael S., Furjen Deng, and Lou-Jou Lee. 2003. “Evaluating a Prison-Based Drug Treatment 
Program in Taiwan.” Journal of Drug Issues 33, (2): 357–83. 
doi:10.1177/002204260303300205.
World Health Organization. Preventing youth violence: an overview of the evidence. World Health 
Organization, 2015.
81
ANNEX I. PROGRAM BACKGROUND
Table 1. Juntos para la Prevención de la Violencia Results Framework
Goal: Reduce violence in target regions
Development objective: Contribute to the reduction of violence and crime by 
strengthening adaptation and implementation capacities of local systems of 
prevention.
Intermediate Result 1. 
Support the Government 
of Mexico (GOM) in 
improving national 
violence and crime 
prevention policy and with 
its collaboration with local 
actors.
Intermediate Result 2. 
Improve the performance 
of actors within local 
violence and crime 
prevention systems.
Intermediate Result 3. 
Contribute to local systems 
generating and implementing 
evidence-based violence and 
crime prevention policy.
Component 1.1. The 
Government of Mexico 
adopts and applies 
continuous training and 
knowledge management 
frameworks to public 
officials on matters of 
prevention.
Component 2.1. The 
local actors have 
improved their 
performance in matter of 
institutional enforcement 
and crime prevention
Component 3.1. Actors 
within local prevention 
systems have improved 
their capacities to produce 
evidence
Component 1.2. The 
Government of Mexico 
adopts and applies tools 
that facilitate the design 
and the implementation of 
the national violence and 
crime prevention policy.
Component 2.2. 
Strategic actors within the 
local system collaborate in 
more efficient ways in 
CVP initiatives
Component 3.2. Local 
actors have increased the 
demand and use of 
knowledge on matters of 
violence and crime 
prevention
82
ANNEX II. DATA COLLECTION INSTRUMENTS
II.1 ONLINE SURVEY
I. General Information
1. Name of organization: 
2. Region:
a. Tijuana, Baja California
b. Chihuahua, Chihuahua
c. Ciudad Juárez, Chihuahua
d. Metropolitan Area of Monterrey, Nuevo León
e. Metropolitan Area of Guadalajara, Jalisco
f. Morelia, Michoacán
3. Sector:
a. Local Government
b. State Government
c. Federal Government
d. Private Sector
e. Civil Society
f. Academia
4. Current Role:
a. Management or Leadership
b. Administration
c. Operational Staff
d. Other: ____________
5. Has your organization received financial assistance from JPV?
a. Yes
b. No
c. N/A
6. Has your organization received technical assistance from JPV? Technical assistance here refers to 
professional or special services to support and strengthen the technical and organizational capacities of 
organizations, including guided instruction, skills training, methodologies and materials development, 
strategic planning, transmission of special expertise, among others.
a. Yes
b. No
c. N/A
7. Does your organization directly implement any project, program, or public policy to prevent 
violence or crime? 
a. Yes
b. No
8. Does your organization support any project, program, or public policy to prevent violence or 
crime by means of financial or technical assistance?
a. Yes
b. No
Please answer the following questions based on your interactions with JPV/USAID over the course of the past four 
years.
83
II. Resources
These questions all pertain to the resources that the JPV/USAID program has provided:
1. Which of the following resources provided by JPV/USAID to strengthen local systems and 
improve program implementation are you familiar with? (Please select all that apply)
a. Systematization of best practices
b. Publications with evidence and research initiatives
c. Tools developed by JPV / USAID (YTT, GOPI -P, OPI-P, EB P)
YTT = Youth Target Tool
GOPI = Government Performance Index
OPI = Organizational Performance Index
EBP = Scale of Best Practices
d. Guides and manuals for the implementation of violence prevention programs
e. Personnel training (diagnosis, evidence-based intervention design, theory of change 
and indicator design, monitoring and evaluation, communication)
f. Study Trips
g. National/International events (conferences, forums, summits, etc.)
h. Diffusion or promotion of best practices
i. Other: __________
2. Which of the following resources provided by JPV/USAID to strengthen local 
systems and improve program implementation do you consistently use at your 
work?
a. Systematization of best practices
b. Publications with evidence and research initiatives
c. Tools developed by JPV / USAID (YTT, GOPI -P, OPI-P, EBP) 
d. Guides and manuals for the implementation of violence prevention programs
e. Personnel training (diagnosis, evidence-based intervention design, theory of change 
and indicator design, monitoring and evaluation, communication)
f. Study Trips
g. National/International events (conferences, forums, summits, etc.)
h. Diffusion or promotion of best practices
i. Other: __________
j. None
3. On a scale of 1 to 5, with 1 being least useful and 5 being most useful, how useful 
would you consider the following resources in informing your daily operations?
a. Systematization of best practices (1 , 2 , 3 , 4 , 5)
b. Publications with evidence and research initiatives (1 , 2 , 3 , 4 , 5)
c. Tools developed by JPV / USAID (YTT, GOPI -P, OPI-P, EB P) 
YTT = Youth Target Tool
GOPI = Government Performance Index
OPI = Organizational Performance Index
EBP = Scale of Best Practices
(1 , 2 , 3 , 4 , 5)
d. Guides and manuals for the implementation of violence prevention programs
(1 , 2 , 3 , 4 , 5)
e. Personnel training (diagnosis, evidence-based intervention design, theory of change 
and indicator design, monitoring and evaluation, communication)
(1 , 2 , 3 , 4 , 5)
84
f. Study Trips (1 , 2 , 3 , 4 , 5)
g. National/International events (conferences, forums, summits, etc.)
(1 , 2 , 3 , 4 , 5)
h. Diffusion or promotion of best practices (1 , 2 , 3 , 4 , 5)
4. If you have not participated in the events or personnel trainings provided by 
JPV/USAID, what is the reason? Please select all that apply. 
a. Events and/or personnel training have not been developed in your region
b. You participated before and materials had not been updated. 
c. You have access to other events and / or trainings through your own organization or 
others.
d. They don’t seem useful
e. They are not necessary for the role that you perform
f. You were not aware of these resources and/or were not made aware of the 
invitation
g. Other: __________
5. Why have you not used the tools and/or publications developed by JPV/USAID?
a. They are difficult to interpret and apply
b. They are not easily accessible
c. There are better tools and/or publications in your own organization, internally or via 
other organizations
d. They don’t seem useful
e. They are not necessary for the role that you perform
f. You were not aware of these resources 
g. Other: __________
6. Which of the following resources do you consider most useful for the 
identification, design, and implementation of best practices? (Select all that 
apply)
a. Systematization of best practices
b. Publications with evidence and research initiatives
c. Tools developed by JPV / USAID (YTT, GOPI -P, OPI-P, EBP)
d. Guides and manuals for the implementation of violence prevention programs
e. Personnel training (diagnosis, evidence-based intervention design, theory of change 
and indicator design, monitoring and evaluation, communication)
f. Study Trips
g. National/International events (conferences, forums, summits, etc.)
h. Dissemination or promotion of best practices
i. Other: __________
7. To what extent do you agree or disagree with the following statements 
related to the technical capacity trainings provided by the JPV/USAID 
program?
a. For the most part, the training material was technically solid.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
85
b. The training materials were relevant for my organization.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
c. JPV/USAID provided the support and follow-up necessary to implement the 
capacities learned in each training. 
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
*Please leave any additional comments here: ___________
8. Would you say that your organization has been able to adopt the capacities 
developed with the support of JPV/USAID?
a. Yes
b. No
c. N/A
9.In what way(s) has your organization adopted or institutionalized the 
capacities developed?
________________________________________________________________
10. Do you consider the resources and supplies provided by JPV/USAID to be 
consistent and predictable enough to plan for their efficient use in the future? 
Consistent here means that there is no time variation and predictable means that you 
can know with certainty that they will be available in the future. 
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
11. As a product of cooperation with JPV/USAID, has your organization 
identified alternative sources for additional financing from other actors￾private sector, government funding, trusts or other- be they local or outside, 
to ensure the sustainability of its prevention initiatives? 
a. Yes
b. No
2. What necessary resources could be provided through international 
cooperation to strengthen local prevention systems that do not currently 
exist or exist in limited fashion?
_______________________________
86
III. Roles
The following questions will pertain to the participation of relevant actors and their roles in the local prevention 
systems: 
1. To what extent would you agree or disagree with the following statements 
regarding the local context?
a. JPV/USAID effectively identified the relevant local actors who work in 
crime and violence prevention in your region. (including the private 
sector, government, academia, and civil society)
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
b. JPV/USAID has a good understanding of the local context and the 
challenges faced by relevant actors.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
c. JPV/USAID’s plan of action and recommendations are supported by a 
thorough understanding of local contexts.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
 
*Please leave any additional comments here: ___________
2. How would you rate the performance of the following actors in preventing 
crime and violence in your region in terms of efficiency? By efficient we refer 
to their capacity to adequately perform or complete their respective role. 
a. State Government
a. Excellent
b. Good
c. Neutral
d. Poor
e. Very poor
b. Municipal Government
a. Excellent
b. Good
c. Neutral
d. Poor
e. Very poor
c. Private Sector
87
a. Excellent
b. Good
c. Neutral
d. Poor
e. Very poor
d. Civil Society
a. Excellent
b. Good
c. Neutral
d. Poor
e. Very poor
e. International Cooperation
a. Excellent
b. Good
c. Neutral
d. Poor
e. Very poor
f. Academia
a. Excellent
b. Good
c. Neutral
d. Poor
e. Very poor
3. Who do you believe have the most influence in policies for crime and 
violence prevention in your region? On a scale from 1 to 5, please indicate 
who you believe to have the most influence, with 1 being least influential and 
5 being most influential. 
a. Civil society (1 , 2 , 3 , 4 , 5)
b. International cooperation (1 , 2 , 3 , 4 , 5)
c. State Government (1 , 2 , 3 , 4 , 5)
d. Municipal Government (1 , 2 , 3 , 4 , 5)
e. Academia (1 , 2 , 3 , 4 , 5)
f. Private sector (1 , 2 , 3 , 4 , 5)
4. Would you say that JPV/USAID involves relevant actors in the activities it 
organizes?
a. JPV/USAID involves all relevant actors 
b. JPV/USAID involves the majority of relevant actors 
c. JPV/USAID involves some relevant actors
d. JPV/USAID involves very few relevant actors 
e. JPV/USAID does not involve any of the general actors 
5. How would you describe the changes to private sector involvement in crime 
and violence prevention has changed since the beginning of the JPV/USAID 
program?
a. The private sector is more involved
b. The private sector is somewhat more involved
88
c. Private sector involvement is unchanged
d. The private sector is somewhat less involved
e. The private sector is less involved/not involved
6. How much influence does civil society have in the local crime and violence 
prevention system?
a. Local system actors always take the opinion of civil society into account. 
b. Local system actors usually take the opinion of civil society into account.
c. Local system actors sometimes take the opinion of civil society into account. 
d. Local system actors rarely take the opinion of civil society into account. 
e. Local actors never take the opinion of civil society into account. 
To what extent would you agree or disagree with the following statements 
regarding the local context?
7. JPV has been able to empower Civil Society organizations to occupy a more 
prominent role in violence prevention.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
8. JPV has facilitated improved communication in the local prevention 
network.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
9. There is clear-cut leadership that can bring together the community within 
the local violence prevention system.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
10. Where does this leadership stem from?
a. Civil Society
b. International Cooperation (JPV/USAID)
c. State Government
d. Municipal Government
e. Academia
f. Private Sector
11. The leaders in the local prevention system are effective in coordinating local 
system actors and using the existing resources.
a. Strongly Agree
89
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
12. Your region has sufficient capacity to effectively address youth at risk of 
involvement in criminal or violent activity. 
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
13. How has this capacity changed since JPV/USAID involvement in your region?
a. It has significantly increased
b. It has increased
c. It has remained the same
d. It has decreased
e. It has significantly decreased
IV. Relationships
The following questions pertain to the communications and relationships between local s:
1. How would you characterize your relationship with JPV/USAID? On a scale from 
1 to 5, please indicate to what degree you would consider each adjective 
applicable, with 1 being not at all and 5 very much. 
a. Hierarchical (1 , 2 , 3 , 4 , 5)
b. Reciprocal (1 , 2 , 3 , 4 , 5)
c. Collaborative (1 , 2 , 3 , 4 , 5)
d. Beneficial (1 , 2 , 3 , 4 , 5)
e. Conflictive (1 , 2 , 3 , 4 , 5)
2. How would you characterize your relationship with similar organizations to 
your own that work in the local prevention system?
a. Very strong
b. Somewhat strong
c. Neither strong or weak
d. Somewhat weak
e. Very weak
3. How would you characterize your relationship with Academia? On a scale from 
1 to 5, please indicate to what degree you would consider each adjective 
applicable, with 1 being not at all and 5 very much. 
a. Hierarchical (1 , 2 , 3 , 4 , 5)
b. Reciprocal (1 , 2 , 3 , 4 , 5)
c. Collaborative (1 , 2 , 3 , 4 , 5)
d. Beneficial (1 , 2 , 3 , 4 , 5)
e. Conflictive (1 , 2 , 3 , 4 , 5)
90
4. How would you characterize your relationship with the municipal government? 
On a scale from 1 to 5, please indicate to what degree you would consider each 
adjective applicable, with 1 being not at all and 5 very much. 
f. Hierarchical (1 , 2 , 3 , 4 , 5)
g. Reciprocal (1 , 2 , 3 , 4 , 5)
h. Collaborative (1 , 2 , 3 , 4 , 5)
i. Beneficial (1 , 2 , 3 , 4 , 5)
j. Conflictive (1 , 2 , 3 , 4 , 5)
5. How would you characterize your relationship with the state government? On a 
scale from 1 to 5, please indicate to what degree you would consider each 
adjective applicable, with 1 being not at all and 5 very much. 
a. Hierarchical (1 , 2 , 3 , 4 , 5)
b. Reciprocal (1 , 2 , 3 , 4 , 5)
c. Collaborative (1 , 2 , 3 , 4 , 5)
d. Beneficial (1 , 2 , 3 , 4 , 5)
e. Conflictive (1 , 2 , 3 , 4 , 5)
f. N/A
6. How would you characterize your relationship with local foundations, if you 
receive funds from one of them? On a scale from 1 to 5, please indicate to what 
degree you would consider each adjective applicable, with 1 being not at all and 
5 very much. 
a. Hierarchical (1 , 2 , 3 , 4 , 5)
b. Reciprocal (1 , 2 , 3 , 4 , 5)
c. Collaborative (1 , 2 , 3 , 4 , 5)
d. Beneficial (1 , 2 , 3 , 4 , 5)
e. Conflictive (1 , 2 , 3 , 4 , 5)
f. N/A
7. If the JPV/USAID program were terminated, how would the spaces for sharing 
ideas/experiences between other local system actors be affected?
8. Would you say that JPV/USAID’s feedback has been valuable?
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
*Please leave any additional comments here ____________
9. Have you had opportunities to give JPV/USAID your feedback?
a. Yes
b. No
Why or why not?
91
10. JPV has taken your feedback into account.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
* Please leave any additional comments here:_______
V. Rules
The following questions pertain to the details of your collaboration with JPV/USAID:
1. What type of contractual relationship do/did you have with JPV/USAID?
a. Subcontract
b. Donation
c. Cooperation Agreement
d. No formal relationship
e. Other: _________
2. Who assumed more of the risk in this contractual relationship?
By risks we refer to financial, programmatic risks, as well as risks to reputation.
Please indicate a position on the line corresponding to who assumed more risk.
_________________________________________________________________
3. Please indicate, from 1 to 5, how relevant you think each criterion was for 
JPV/USAID to determine which organizations would receive a donation. On a 
scale from 1 to 5, please indicate how relevant would you consider each 
criterion, with 1 being not at all relevant and 5 very relevant. 
a. Evidence-based practices (1 , 2 , 3 , 4 , 5)
b. Implementer’s organizational capacity (1 , 2 , 3 , 4 , 5)
c. Implementer’s innovative capacity (1 , 2 , 3 , 4 , 5)
d. Low risk associated with the donation (1 , 2 , 3 , 4 , 5)
4. The contractual terms of the relationship with USAID allowed for the 
flexibility necessary to implement your program.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
f. N/A
* Please leave any additional comments here:________________________________________
JPV/USAID Risk equally 
shared between 
parties
My 
organization
92
5. The adjustments made to your program design as a result of cooperation with 
JPV/USAID improved the effectiveness of its implementation. 
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
f. N/A
*Please leave any additional comments here:_______
6. Who had more decision-making power throughout the implementation of the 
program? Please indicate a position on the line corresponding with who had more 
decision-making power.
_________________________________________________________________
7. Implementation plans commonly have to make adjustments as public policy 
advances. In your opinion, how receptive were JPV/USAID to the necessity of 
these adjustments?
a. JPV/USAID was very receptive and facilitated the necessary changes
b. JPV/USAID was somewhat receptive and facilitated some of the necessary changes
c. JPV/USAID was little receptive, and facilitated few of the necessary changes 
d. JVP/USAID was not receptive.
8. How would you classify the reporting requirements on the progress of your 
program to JPV / USAID? 
a. Minimal
b. Less than sufficient
c. Sufficient
d. More than sufficient
e. Excessive 
*Please leave us a comment regarding your perception of the reporting requirements: _______
9. The indicators you must report to JPV/USAID are:
A. Useful
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
B. Reflective of the advances made by the program
a. Strongly Agree
b. Agree
JPV/USAID Same decision￾making power
My 
organization
93
c. Neutral
d. Disagree
e. Strongly Disagree
C. Relevant
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
2. The time dedicated to completing the required reports for USAID was: 
a. Very useful
b. Somewhat useful
c. Neither useful or useless
d. Somewhat useless
e. Very useless 
*Please leave any additional comments here:___________
2. In your opinion, the objectives established in the beginning of your 
collaboration with JPV/USAID were: 
a. Pessimistic
b. Somewhat pessimistic
c. Neither pessimistic or optimistic
d. Somewhat optimistic
e. Very optimistic
3. You had to commit yourself to some unrealistic goals in order to receive
funding from JPV/USAID.
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree
4. The implementation calendars committed to JPV are realistic 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
VI. Results
The final questions pertain to the results that the program with JPV/USAID has achieved:
94
1. To what extent do you agree or disagree with the following statements 
related to local crime and violence prevention systems? 
A. JPV/USAID contribution led to local system actors identifying, designing, and 
implemented programmatic high-potential prevention policies 
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
B. JPV/USAID contributed to local actors in the violence prevention system 
assigning more resources to effective prevention strategies. 
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
C. JPV/USAID contributed to a diverse set of actors being more sensible about the 
importance of crime and violence prevention.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
D. JPV/USAID made it feasible for all actors to participate and work as a local system.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
2. On a scale from 1 to 5, with 1 being very little change and 5 being drastic 
change, how much do you consider JPV/USAID was capable of positively 
influencing the institution below in their crime and violence prevention 
work? 
a. State government (1 , 2 , 3 , 4 , 5)
b. Municipal government (1 , 2 , 3 , 4 , 5)
c. Civil society (1 , 2 , 3 , 4 , 5)
d. Academia (1 , 2 , 3 , 4 , 5)
e. Private sector (1 , 2 , 3 , 4 , 5)
* Please leave any additional comments here: _________________________
3. To what extent do you agree or disagree with the following statements 
related to the role of your organization as it compares to JPV/USAID and 
other system actors? 
A. JPV/USAID made it easier for your agency to assume a local leadership role vis-à-vis
other actors. 
a. Strongly Agree 
b. Agree 
c. Neutral 
95
d. Disagree 
e. Strongly Disagree 
B. JPV/USAID made it easier for your agency to gain local recognition vis-à-vis other 
actors. 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
C. JPV/USAID made it easier for your agency to implement the practices, tools, and 
initiatives developed in conjunction with JPV/USAID. 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
D. JPV/USAID made it easier for your organization to develop a broader network of 
contacts that work in crime and violence prevention. 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
4. Has your organization managed to have influence on local regulations 
regarding crime and violence prevention and/or has influenced in the local 
government proposed public policy during the last 4 years? Influence here refers 
to actions that organizations have taken with the intention of directly altering political decision, 
actions, and policies that are designed, adopted, or implemented by Governmental 
a. Yes
b. No
c. N/A. 
Please explain:__________________
5. Do you know about any organization besides yours that has influenced in 
local violence prevention regulations and/or adoption of proposed public 
policies during the last four years? 
a. Yes (If so, which one?)
b. No
6. Would you say that JPV/USAID created the conditions for your organization
an impact on public policies for crime and violence prevention?
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
96
7. To what extent do you agree or disagree with the following statements 
related to the implementation of best practices? 
A. JPV/USAID contribution led to implementing organizations from the local crime and 
violence prevention system to elaborate diagnosis with disaggregated data that take 
sociodemographic characteristics of the region and risk factors of the population 
being served.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
B. JPV/USAID contribution provided implementer organizations with more tools to 
identify and follow up with the beneficiaries of the interventions.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
C. JPV/USAID contributed to implementer organizations' improved capacity and more 
tools to monitor and evaluate the success of interventions.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
D. JPV/USAID contributed to implementer organizations implementing best practices for 
prevention in a sustainable way.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
E. JPV/USAID contributed to local organizations implementing prevention strategies that 
will have continuity in the long run.
a. Strongly Agree
b. Agree
c. Neutral
d. Disagree
e. Strongly Disagree
* Please leave any additional comments here:_________________________
Only a few questions left. 
8. To what extent do you agree or disagree with the following statements 
related to local innovation and transference of best practices? 
A. JPV/USAID made it easier for local system actors to access best practices materials 
developed by other actors. 
a. Strongly Agree 
97
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
B. JPV/USAID contribution led to local systems catalyzing innovations at a local level. 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
C. JPV/USAID made the transfer of knowledge between actors easier. 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
¿ Which innovations you know produced at a local level have been replicated 
elsewhere or scaled by other actors? ______________________
9. To what extent do you agree or disagree with the following statements 
related to monitoring and evaluation processes? 
A. JPV/USAID contributed better indicators for monitoring the progress of the 
implemented programs' activities. 
a. Strongly Agree 
b. Agree
c. Neutral 
d. Disagree 
e. Strongly Disagree 
B. JPV/USAID made it easier for your organization to develop the capacity to carry out 
monitoring and evaluation by themselves. 
a. Strongly Agree
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
C. JPV/USAID facilitated cooperation with organizations of process evaluation experts. 
a. Strongly Agree 
b. Agree 
c. Neutral 
d. Disagree 
e. Strongly Disagree 
10. If you did not have support from USAID, what would you consider to be the principal 
challenges facing your organization? Why?
11. Notwithstanding support from USAID, what would you consider to be the principal 
challenges facing the local prevention system? Why?
98
12. Would you say that the local prevention system is adaptable and capable of responding to 
political changes? Why or why not?
13. What could the program have done differently/ what would you recommend for future 
initiatives?
99
II. 2 GENERAL INSTRUMENT FOR SEMI-STRUCTURED INTERVIEWS 
(RELEVANT STAKEHOLDERS)
Initial Questions:
1. To begin, we would like to hear more about your experience working with JPV/USAID. When 
did you begin collaborating with them? How closely did you work together?
2. Which aspects of your collaboration with JPV/USAID were the most positive? What has worked 
well?
3. Which aspects of your collaboration could be improved? What has not worked well?
4. What are the principal limitations the JPV/USAID confronted in the local prevention system?
Questions pertaining to local prevention systems:
I. Resources
1. Did you receive resources or materials from JPV/USAID? (Y/N) 
i. Do you consider the resources and supplies provided by JPV/USAID to be 
consistent and predictable enough to plan for their efficient use in the future? 
Why or why not?
ii. Which resources were unreliable or scarce and how did your organization 
confront this challenge?
2. Did your organization receive capacity training from JPV/USAID? (Y/N)
i. Was the follow-up to the trainings appropriate and sufficient to implement the 
capacities learned in the training(s)?
ii. In what way(s) has your organization adopted or institutionalized the capacities
developed as to not depend on the resources of international cooperation?
iii. What challenges does your organization face in accessing resources for capacity 
training in the absence of JPV/USAID support?
II. Roles
1. Could you tell us more about the relationship between the government, civil society and 
the private sector in your region? Has this changed in light of the JPV/USAID program? If 
so, in what way? What are the communication channels that you rely on?
2. Have JPV/USAID been effective in identifying relevant actors in the local system? What 
were some of JPV/USAID’s major successes/failures in identifying/involving these relevant 
actors?
3. Is there a clear leadership in organizational authority in the local system? Who is in charge 
of this? Is it adequate? Why or why not?
4. Would you say that JPV/USAID have been capable of empowering relevant actors to take 
more prominent roles in violence prevention?
i. For example, would you agree that selecting _____ as a key actor/ strategic 
partner was a success? [This question changes for each region]
5. Since the beginning of the JPV/USAID program, which actors have increased their 
participation in the local violence prevention system the most? What kind of activities are 
they involved in?
i. Does your organization have the power/influence to impact the direction of the 
local prevention system?
ii. Which organizations have the most influence in the local prevention system? 
Why?
100
III. Relationships
1. Did the priorities of the JPV/USAID program take the needs and aspirations of yours and 
other local organizations into account?
i. How do their priorities change in response to the demands of local actors and 
how do they adapt to them? 
2. Was there adequate space for giving JPV/USAID feedback from your organization 
regarding your collaboration with them? Did they take that feedback into account?
i. How receptive did you perceive the program to be? Were there perceptible 
differences between different staff positions?
3. Have JPV/USAID been successful in bringing about positive interactions between local 
actors where before there were none?
i. What are JPV/USAID’s most important achievements in this respect? Which were 
the most important shortcomings?
ii. Has your organization extended its network of organizations and contacts as a 
result of the JPV/USAID program?
4. Do you consider there are instances for accountability within the local prevention system? 
IV. Rules
1. What kind of contractual relationship did your organization have with JPV/USAID? Was 
it appropriate? Why or why not?
2. How realistic were the goals your organization committed to in order to receive 
financial/technical assistance from JPV/USAID? Please elaborate on JPV/USAID’s 
successes/failures on in developing the contract/program with your organization.
3. In what way did JPV/USAID’s proposed time horizons impact your organization’s work?
4. Was your organization given the flexibility needed to make efficient use of JPV/USAID’s 
resources? Why or why not?
5. How much risk did USAID assume in deciding to open its investment portfolio to 
organizations that could act as catalysts for promising operations? 
i. What was balance between risk/reward in terms of returns?
ii. Which criteria were most relevant for JPV/USAID to award/renew its donations 
to your organization?
V. Results
1. Has your organization been strengthened since its cooperation with JPV/USAID? In what 
way have donations/technical assistance contributed to this end?
2. Do you think that your organization through has more tools and capacity to identify and 
design effective and promising models informed by evidence as result of collaboration 
with JPV/USAID?
i. How did JPV/USAID contribute to this result?
3. Was JPV/USAID effective in conducting knowledge transfer between actors in different 
regions?
i. Were the results generated by other actors (implementation guides, resources, 
publications, research, etc.) effectively translated into usable materials for your 
organization?
4. Has JPV/USAID been effective in transferring effective models between actors?
i. In particular, what do you believe accounts the success of models that local 
actors have explicitly requested? (Ex. CBT o Civic Justice)
5. What innovations generated by other actors (national or international) did your 
organization have access to? 
101
i. What do you consider to be the advantages and disadvantages to focusing on 
funding institutions that can act as catalysts for contextualized innovations that 
are scalable in the long-term as opposed to adopting models of famous foreign 
franchises?
6. In the absence of support from JPV/USAID, what are the principal challenges that the local 
prevention system faces? Why?
i. Is the local prevention system capable of adapting and responding to political 
changes? Why or why not?
7. In the absence of support from JPV/USAID, what are the principal challenges that your 
organization faces? Why?
8. From your experience with JPV/USAID, what role do you think international cooperation 
in crime and violence prevention should occupy? Which activities should it direct its 
limited resources to? 
i. What resources should be provided by international cooperation that are 
necessary for the implementation of best practices in prevention? (Note for 
interviewer: We are seeking to understanding more about resources that the local system 
has no substitutes for).
9. How have JPV/USAID contributed to the reduction of crime and violence in the region? 
How have they achieved results?
102
II. 3 INSTRUMENT FOR SEMI-STRUCTURED INTERVIEWS WITH CSOS 
PROGRAM COORDINATORS 
I. Targeting
a. What is the target population of your organization?
i. Which mechanisms or strategies do you use to reach this population? Which are 
the reasons for rejecting a case?
ii. Which tools do you utilize to determine youth's profiles? What do you measure 
with these instruments? How did you choose or develop these instruments?
(inquire if these are instruments that they built and validated or if they took them 
from somewhere else)
iii. How did JPV/USAID contribute to improving these profiling tools? (YTT, 
generation of disaggregated crime incidence data by neighborhoods, etc.)?
b. Which risk factors you are trying to mitigate with your interventions with youth?
i. How effective do you think your interventions are in achieving this goal? Why?
ii. How did JPV/USAID help your organization to better understand and identify risk 
factors and local dynamics of violence? Which resources and/or tools did 
JPV/USAID provide to your organization to better understand and identify risk 
factors and local dynamics of violence?
c. Which are the protective factors that you seek to strengthen in young people?
i. How do you enhance them?
ii. How did JPV/USAID help your organization to better understand and identify 
youth’s protective factors?
II. Attention to young people and families
a. Which services does your organization offer to young people and their families? Which 
of them are under your coordination?
i. Which changes have you made since the beginning of the collaboration with JPV?
ii. In your area, how did you define this offer of services? How do you update it?
How often?
iii. Which opportunities for improvement do you identify in working with young 
people in the area you coordinate?
iv. How do you work with families?
b. How do you determine the activities in which each youth will participate?
i. Which are the main difficulties you face in engaging young people with these 
activities? What about engaging families?
ii. Which strategies do you follow when youth lack of commitment? What helps 
youth engage and stay in the program?
c. Which are the tools you offer young people to reduce their recidivism likelihood after
they finish the program in your organization?
i. How do you determine when a youth is ready to leave the program?
ii. How often do you have youth who had already been with you before?
III. Implementation capabilities
a. With which government institutions does your organization collaborate with? 
i. How did JPV/USAID facilitate coordination with these institutions?
ii. How are the dynamics of work with these institutions? How does this dynamic 
affect the work with youth?
b. Specifically, in the case of educational institutions, which challenges do you identify to 
ensure that youth continue with their education?
103
i. How do educational modalities and study plans help this youth to keep up with 
their studies?
c. Which companies and private sector organizations are you working with you?
i. How do companies or private sector organizations support youth’s reintegration 
processes? What motivates companies or private sector organizations to work 
with young people?
ii. How effective are linkages with these companies?
iii. How often do they offer formal job opportunities to youth? Are offered wages 
sufficient?
d. In which levels of prevention does your organization work? Which are your organization’s
capabilities to implement strategies at all prevention levels?
i. How has JPV/USAID contributed to strengthen these capacities?
ii. Which capacities remain to be generated to implement the strategies more 
effectively? Which are the biggest challenges, in terms of capabilities, that your 
organization faces to effectively implement these strategies? (human capital, inter￾institutional financial coordination or technical knowledge)? 
iii. How did JPV/USAID contribute to meet these challenges
e. Which actors have been involved in the design, implementation, and evaluation of your 
organization's interventions?
i. Did JPV/USAID strengthen cooperation with other specialized actors in the 
design, implementation, and evaluation of interventions?
ii. Which relationships are yet to be developed with key stakeholders to more 
effectively implement prevention strategies?
f. Do you consider that your organization has the necessary resources to effectively carry 
out its work?
i. How has JPV/USAID contributed to guaranteeing sustainable access to these 
resources?
ii. Which additional resources would be necessary to carry out the implementation 
more effectively?
IV. Monitoring and evaluation
a. How do you follow-up on cases after program conclusion? (Probe on family follow-up)
b. Which monitoring tools do you use in your organization to follow-up on interventions’
implementation?
i. How have these tools changed and/or been strengthened through cooperation 
with JPV / USAID?
c. How do you evaluate the impact and/or success of interventions in your organization?
i. How have measurement/evaluation tools changed and/or been strengthened 
through cooperation with JPV / USAID?
d. Do you consider that due to the work done with JPV/USAID, the perception of the 
importance of evaluating and informing by evidence the work you do, has changed within 
your organization?
V. Institutionalization
a. Is/are the intervention(s) implemented by your organization institutionalized?
b. Which factors pose a threat to the sustainability and institutionalization of your 
interventions? (change of government, lack of financing, etc.)?
c. In your organization, have you considered expanding to other contexts or transferring 
the practice to other actors?
d. How has JPV/USAID enhanced the scale potential of your interventions
104
ANNEX III. RESULTS ACHIEVED BY MEASSURMENT INSTRUMENT
Table 1. GOPI-P Scores for JPV Supported Municipalities
Municipality 1st
measurement
(2016-17)
2nd
measurement
(2018)
3rd
measurement
(2019)
% 
Change
Pending 
improvement areas
Chihuahua 37% Willing to 
Participate
54% Basic 69% Basic 86% - Design a 
communication strategy 
for C&V prevention
Ciudad 
Juárez
41% Willing to 
Participate
59% Basic 77% Acceptable 88% - Institutionalize the 
multisectoral group - 
Mesa del Suroriente, in 
the municipal 
government
Escobedo 48% Willing to 
Participate
68% Basic 65% Basic 35% - Improve citizen 
involvement 
- Strengthen use of C&V 
prevention tools 
- Update crime and 
violence prevention 
strategy
Guadalupe 57% Basic 64% Basic 52% Basic - 9% - Train public servants 
on strategic planning
- Reestablish 
multisectoral group
Monterrey 33% Willing to 
Participate
69% Basic 84% Acceptable 155% - Establish the Municipal 
Council for C&V Social 
Prevention 
- Trainings 
Morelia 38% Willing to 
Participate
N/A74 58% Basic 53% N/A
Tijuana 51% Basic 65% Basic 65% Basic 27% - Improve the 
multisectoral group’s 
structure
- Develop specialized 
studies on C&V
Tonalá 53% Basic 60% Basic 52% Basic -2% - Improve citizen 
involvement
- Develop specialized 
diagnoses
- Generate a 
communication strategy
Zapopan 59% Basic 57% Basic 69% Basic 17% - Improve M&E capacity
- Build multisectoral 
alliances 
- Design V&C 
prevention public policy
74 The second meassurment of the GOPI-P in Morelia was focused on Civic Justice and not in the overall crime and violence 
prevention agenda, and therefore it is not considered.
105
Table 2. OPI-P Results for Organizations that Received Strengthening 
Assistance Plan
Municipality CSO Initial OPI Final OPI Change
Chihuahua Busuleba A.C. Low Low -
Centro de Liderazgo y Desarrollo 
Humano
Basic N/A -
Fomentando bienestar y cohesión 
comunitaria – Programa Espere
Low Basic +
Red de Cohesión Social 1,2,3 por 
mí y mi comunidad**
High High -
Ciudad Juárez Casa Triunfo de México, A.C.* Low Low -
Sin Violencia A.C. Low N/A -
Techo Comunitario Basic Moderate +
Guadalajara 
(metropolitan 
region)
Aluzarte Basic Basic -
Corpocreativo cultura en 
movimiento A.C.
Basic Moderate +
México Me Necesita A.c. Moderate N/A -
Paz y Reconstrucción del Tejido 
Social A.C.
Basic Moderate +
Seguimiento y Detección del 
Intento Suicida y Abuso Sexual 
A.C. (SYDISAS)
Moderate Moderate -
Monterrey 
(metropolitan 
region)
Alternativas Pacíficas A.C. Moderate Moderate -
Despierta, Cuestiona y Actúa A.C. Moderate N/A -
Fundación Jomar Low Basic +
Mariposas de la Felicidad Basic Basic -
Raza Nueva Low Basic -
Pastoral Juvenil en Situaciones 
Críticas PJSC A.C.
Low Basic -
Morelia Centros de Integración Juvenil* Basic N/A -
Fundación Crece Low N/A -
Tijuana Casa de las Ideas Basic Moderate +
Centro de Arte y Cultura El Globo
A.C.
Basic Moderate +
Fundación Tú + Yo A.C. Basic
Moderate75
Moderate +
Promotora de las Bellas Artes A.C. Basic Moderate +
* Organizations that received funding from JPV. 
** Organizations that also have a Best Practice Scale (EBP) and received funding from JPV
75 This CSO has different results reported in DevResults and in its report. The results are Basic in the former and Moderate in 
the Later. 
106
Table 3. Best Practices Score (EBP, for its acronym in Spanish)
Municipality Name of 
intervention
Implementing 
organization
Initial EBP 
Category
Final EBP 
Category
Change Received 
JPV 
funding
Chihuahua
Modelo de familias 
inteligentes CIFAC
Not informed 
by evidence
Promising 
practice
+ -
Mi Colonia es mi Casa
Instituto Municipal 
de Planeación 
(Comisión CPTED 
Urbano)
Informed by 
evidence
Promising 
practice
+ -
Comisión de Juventud 
en Riesgo
Comisión de 
Juventud en Riesgo
Informed by 
evidence
Promising 
practice
+ -
Modelo de intervención
comunitaria, 1, 2 y 3 
por mí y por mi 
comunidad
Red de Cohesión y 
Participación Social
Informed by 
evidence
Promising 
practice
+ Yes
Ciudad Juárez
Autonomía y libertad 
interior transformando 
sujetos violentos CASA
Not informed 
by evidence
Informed by 
evidence
+ Yes
Pandillas: Clubs de 
Convivencia Armónica
Sumando Esfuerzos 
por Juárez
Informed by 
evidence
Promising 
practice
+ -
Yo Soy Rediseño Social
Centro Caritativo 
para Atención a 
Enfermos de Sida, 
A.C.
Not informed 
by evidence
Promising 
practice
+
Yes
Desafío: Modelo de 
Formación Juvenil con 
Enfoque en la 
Reducción de Riesgos
Fundación 
Comunitaria de la 
Frontera Norte, A.C.
Informed by 
evidence
Promising 
practice
+ Yes
Casa Equidad*
Casa Amiga
Not informed 
by evidence
Informed by 
evidence
+ -
Cure Violence / Del 
Barrio a la Comunidad*
Centro Caritativo 
para Atención a 
Enfermos de Sida, 
A.C.
Informed by 
evidence
Promising 
practice
+
Cuauhtémoc 
(Mexico 
Federal 
District)
Atención a adolescentes 
en conflicto con la ley Reintegra, A.C.
Not informed 
by evidence
Promising 
practice
+ Yes
Jóvenes en conflicto con 
la ley: acción, formación 
y ocupación Acciona, A.C.
Informed by 
evidence
Promising 
practice
+ Yes
Escuela de Cultura 
Urbana
La Victoria 
Emergente
Informed by 
evidence
Promising 
practice
+ Yes
Guadalajara
(metropolitan 
region)
Grupos de ayuda mutua
Instituto de las 
Mujeres Guadalajara
Informed by 
evidence
Informed by 
evidence
- -
Servers Grupo RE
Informed by 
evidence
Informed by 
evidence
- -
Mindfulness CEDAT
Informed by 
evidence
Promising 
practice
+ Yes
Arte en prisión Aluzarte
Informed by 
evidence
Informed by 
evidence
- -
Más allá de los miedos Paz y Reconstrucción
Informed by 
evidence
Promising 
practice
+ -
107
Table 3. Best Practices Score (EBP, for its acronym in Spanish)
Municipality Name of 
intervention
Implementing 
organization
Initial EBP 
Category
Final EBP 
Category
Change Received 
JPV 
funding
Monterrey
(metropolitan 
region) Renace Jóven Renace ABP
Not informed
by evidence
Promising 
practice
+ Yes
Tijuana
Modelo de Abordaje 
Comunitario e Integral 
Tú + Yo Fundación Tú + Yo
Informed by
evidence
Promising 
practice
+ -
Orale ProSalud
Informed by 
evidence
Promising 
practice
+ Yes
Casa de las Ideas Casa de las Ideas
Informed by 
evidence
Informed by 
evidence
- -
Club de niños y niñas
Club de niños y 
niñas
Not informed 
by evidence
Informed by 
evidence
+ -
Modelo de Desarrollo 
Comunitario Grupo Ruba
Not informed 
by evidence
Informed by 
evidence
+ -
108
ANNEX IV. COMPARATIVE MUNICIPALITIES MATRIX
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
Institutionali￾zed 
multisector 
collaboration
Council for 
Crime and Violence 
Social Prevention of 
the Municipality of 
Chihuahua
(COPREV).
Operates with little 
oversight from JPV.
In 2019, Mesa 
Suroriente was 
established under the 
leadership of the 
Municipal Women’s
Institute to work on 
topics of gender 
violence. The anchor 
actor and JPV 
supported this initiative 
to bring the multisector 
perspective. The 
anchor actor finances 
many of the 
participating 
organizations. In this 
group, members of the 
Executive Committee 
of the Municipal 
Cabinet of Social 
Prevention also 
participate.
76
There is an additional 
working group for 
topics of child sexual 
violence prevention 
A metropolitan 
violence prevention 
committee was 
promoted in 2019 with 
support from the anchor 
actor and in alliance with 
the Metropolitan 
Security Agency and the 
University of 
Guadalajara (this group 
was formed for the 
implementation of civic 
justice).
At the state level, the 
Jalisco Attorney General 
promoted the creation 
of a Youth Reintegration 
Committee (members of 
the committee include 
the government of 
Jalisco, the municipalities
of Zapopan, Guadalajara,
and Tonalá, the Office of 
the Attorney General,
and Corporativa de 
Violence 
Prevention 
Multisectoral 
Network.
A varied set of actors 
participate in this 
agenda, but 
coordination and 
articulation remain a 
challenge. 
Escobedo has its own 
Multisectoral Cabinet 
for the Prevention of 
Violence.
The local system 
is organized around 
the implementation of 
the civic justice. 
Although this is not a 
multisectoral broad 
committee, we deem
this to be sustainable 
over time. 
Municipal 
Committee for 
Violence 
Prevention. 
Operates with 
little oversight 
from JPV, but is 
highly dependent 
on JPV’s funding.
This committee 
was formed since 
the PCC program. 
After the 
administration 
change in 2019, 
giving continuity to 
the work with 
authorities was 
deemed complex. 
Therefore, 
members kept 
meeting under the 
LPS umbrella (also 
adding additional 
relevant actors). In 
2020, the 
76 This committee existed since the former Program for Community Violence Prevention (PCC for its acronym in Spanish) and responded to the PRONAPRED logic. JPV tried to 
reactive this cabinet, but priorities were different. Nevertheless, this committee acts as a counterpart for the GOPI-P application. 
109
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
with participation of 
representatives from 
the local legislature and 
SIPINNA,77 the anchor 
actor, and other local 
organizations.
The Social 
Reintegration 
Committee is an 
additional relevant 
group. This started 
under the leadership of 
a court magistrate and 
in collaboration with 
local NGOs. One 
organization that is 
funded by JPV and the 
anchor actor leads this 
group. Given the 
collaboration, the 
anchor actor has also 
started systematizing 
the work of this 
initiative. 
Fundaciones, the local 
anchor actor).
The municipalities of 
Tonalá and Zapopan also 
achieved 
institutionalization of 
multisector groups.
collaboration with 
the government 
has been slowly 
resumed.
Shared and 
agreed 
agenda and 
clear working 
plans
The group has
been able to 
establish priorities 
and goals around a 
crime-and-violence￾prevention agenda, 
Only as related to 
gender strategy. 
The Mesa Suroriente
agenda is currently 
being developed.
The recently 
established prevention 
committee has made 
progress in defining 
some objectives. It is still 
unclear whether this 
have become actionable.
While the 
multisector group 
was able to attract 
relevant actors, it 
could not coordinate 
and articulate the 
work under a 
Only around the 
implementation of 
civic justice
The group has
been able to 
establish priorities 
and goals around a 
crime-and￾violence￾prevention agenda.
77 National System for the Protection of Girls, Boys, and Adolescents (SIPINNA for its acronym in Spanish)
110
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
organized in five
strategic areas. 
COPREV has an 
internal regulations 
and legal framework.
Zapopan and Tonalá
have developed crime 
and violence agendas 
within each municipality.
common agenda and 
define priorities. 
Escobedo’s cabinet 
meets regularly, but 
it is unclear whether 
setting a clear agenda
(beyond the
implementation of 
civic justice) has been 
possible.
Mesa Metrópoli is 
coordinating a 
metropolitan 
network to 
implement civic
justice in eight 
municipalities of the 
region.
Active 
government 
involvement 
in the 
multisector
group
Although recently 
created, the Mesa 
Suroriente focused on 
the gender strategy is 
fully supported by the 
local government. 
This group was 
recently created. It is 
unclear whether the 
government
participation is 
sustainable. However, 
the Metropolitan 
Security Agency is 
leading the civic justice 
implementation, which is 
promising.
In Escobedo the
participation is active
In Monterrey the 
participation is mainly 
at a state level 
(Nuevo León) 
through the state
attorney.
In Guadalupe there is 
no participation.
After the 
administration 
change in 2019, 
government 
participation was
significantly 
reduced.
111
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
Local system 
impact in 
public policy 
(both through 
the 
multisector 
group or 
outside of it)
COPREV resulted 
in the alignment 
between the 
municipal 
development plan 
and the municipal 
prevention plan, 
ensuring coherence 
in public policy 
regarding crime and 
violence prevention
In 2019, the 
municipal 
government 
approved a civic 
justice regulation to 
substitute the 
municipal edict.78
This substitutes 
some penalties, from 
punitive actions—
such as fines or 
detentions—to 
alternative measures 
that promote the 
restauration of 
community relations. 
The Municipal DIF79
and the Municipal 
Women’s Institute
adopted the Care 
Model for Child and 
Adolescent Victims of 
Family Violence. With 
this model, these 
institutions incorporate 
an evidence-based 
strategy that allows 
them to 
methodologically serve 
children, adolescents, 
and mothers who have 
been exposed to 
violence.
The government of 
Jalisco institutionalized 
the ROLE model for its 
implementation in
private schools.
The government of 
Zapopan, after seeing 
the benefits from the 
Zapopan Rifa project, 
published program
operating rules for its 
institutional 
permanence.
The attorney general 
adopted the 
implementation of a 
CBT-informed 
intervention to support 
youth undergoing 
processes with Jalisco’s 
adolescent criminal 
justice system
Monterrey’s 
government 
partnered with 
RENACE to provide 
psychosocial support 
to youth detained for 
crimes committed in, 
or by people living in, 
Monterrey.
The government of 
Morelia included 
USAID’s local systems 
framework in its 
municipal 
development plan for 
2018–2019
The program 
supported the
development of the 
Directorate for the 
Execution of 
Administrative 
Sanctions that gives 
the institution the 
power to bring 
together other actors 
in the LPS. In this 
capacity, the 
directorate has 
enabled the catalog of 
alternative solutions 
to the conflict that 
will form part of the 
civic justice system. 
The program 
supported the 
development of the 
The government 
approved a reform 
to the municipal 
edict in matters of 
family and gender 
violence.80
The LPS 
independently 
proposed and got
approved the 
integration of a 
crime-and￾violence￾prevention subject 
in all school levels 
of the municipality.
The Violence 
Prevention 
Municipal 
Committee was 
selected to 
administer federal 
funds to finance 
projects for at-risk 
youth.
The anchor actor 
settled an 
agreement with 
78 Bando de Policía y Buen Gobierno in Spanish. 79 National System for the Family Integral Development (DIF for its acronym in Spanish) 80 This regulation now forces violence perpetrators to attend workshops on new masculinities delivered by the Crime Prevention Department of the Municipal Public Security 
Secretariat in alliance with the Municipal Women’s Institute.
112
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
The government 
stated in its 
municipal 
development plan 
for 2019–2021 that 
it would expand the 
program “My 
Neighborhood, My 
House” because of
its effectiveness.
An adapted YTT was 
adopted by the 
criminal justice
system for 
adolescents in the 
state of Chihuahua.
Offender Monitoring 
System (SIMOPIN). 
The objective of this 
system is to generate 
criminals from 
statistical data 
brought to the civic 
courts to inform 
decision-making and 
generate evidence on 
the effectiveness of
the programs that are 
part of the catalog of 
solutions in Morelia.
UVAQ and Morelia’s 
municipal security 
commission formed a 
partnership that aims 
to promote the civic 
justice model and 
provide alternative 
measures for conflict 
resolution. In the 
framework of such 
alliance, UVAQ 
offered to finance 
60% of tuition fees 
for police officers and 
their families for 
undergraduate and 
graduate programs. 
Additionally, UVAQ 
offered stress
management 
workshops for police 
the local 
government to 
operate public 
resources from 
FORTASEG to 
implement the 
project “Youth 
Building 
Prevention,” 
targeting at-risk 
youth, valued at 
4,800,000 MXN 
(about 251,300
USD).
113
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
officers and civic 
judges and 
participated in the 
provision of an 18-
hour-long seminar for 
Morelia’s Municipal 
Security Commission 
and the Municipal 
Receivership.
Private sector 
participation 
in the 
multisector 
group
FICOSEC is also 
an anchor actor and 
co-leads the group.
FICOSEC is 
involved in all of the 
multisector 
collaborations, and 
FECHAS is 
incrementally increasing 
its participation in the 
crime-and-violence￾prevention agenda. 
The private 
sector participation is 
active in Monterrey, 
but not necessarily 
aligned with the LPS
Through the Pro￾Escobedo trust, 
businessman seek to 
contribute to the 
civic justice model by 
generating 
employment for 
offenders in 
companies that 
belong to this 
association. 
COPARMEX The anchor 
actor is financed 
by businessman. 
Other 
organizations from 
the private sector 
are active in the 
group (BBVA, 
Grupo Ruba, etc.) 
but do not 
necessarily finance 
its activities.
Anchor actor 
interested in 
and capable fn 
furthering the 
LPS
Seguridad y Justicia, 
FICOSEC, and 
FECHAC are all 
interested in this 
agenda. It is unclear 
whether the work will
continue the local 
It is unclear 
whether the work 
will continue the 
local system as
defined by JPV.
(The municipal 
government)
But faces a 
financial 
sustainability risk
114
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
system as defined by 
JPV, but there are 
natural leaders for this 
agenda.
Number of at￾risk youth 
direct 
beneficiaries 
of 
reintegration 
programs
Total: 108
Male: 76
Female: 32
Total: 939
Male: 548
Female: 391
Total: 106
Male: 97
Female: 9
Total: 554
Male: 416
Female: 138
Total: N/A
Male: : N/A
Female: : N/A
Total: 316
Male: 171
Female: 145
Total co￾investments81
11,234,201 MXN
(59% was invested in 
CPTED, 34% in 
sexual violence, and 
the rest in at-risk 
youth and civic 
justice)
9,451,764 MXN
(67% was invested in 
at-risk youth, 23% in 
sexual violence, and 
11% in civic justice)
32,880,863 MXN
(75% was for CPTED, 
21% for at-risk youth,
and the rest for civic
justice and a co￾investment from the 
anchor actor)
13,136,413 MXN
(88% was for civic 
justice and 12% for 
at-risk youth. There 
is also a little co￾investment from the 
anchor actor that 
represents less than 
1%)
19,731,789 MXN
(100% was for civic 
justice)
6,346,613 MXN
(88% for at-risk 
youth, 6% for 
gender violence,
and the rest co￾investments from 
the anchor actor 
and other small 
donations)
Public sector 
co￾investment82
6,837,193 MXN
(93% was invested in 
CPTED, the rest in 
civic justice)
1,000,000 MXN83
(100% was invested in 
civic justice)
31,063,632 MXN
(80% of this resource 
was in the municipality 
of Zapopan for CPTED, 
17% for at-risk youth, 
and the rest for civic 
justice)
11,537,790 MXN
(100% was for civic 
Justice)
19,363,222 MXN
(100% was for civic 
justice)
4,891,956 
MXN84
(100% for at-risk 
youth)
81 There is an additional co-investment of 21,519,181 MXN that happened outside the regions, mainly from actors based in Mexico City. 82 There is an additional investment of 18,251,943 MXN coming from FORTASEG made to other regions. 83 FORTASEG resources 84 FORTASEG resources
115
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
Private sector 
co￾investments85
3,977,908 MXN
(96% was invested in 
sexual violence and 
the rest in CPTED 
and family violence)
1,800,000 MXN
(100% invested in 
sexual violence)
26,399 MXN
(100% invested in at-risk 
youth)
- 63,568 MXN
(100% in civic justice)
981,636 MXN
(59% invested in 
at-risk youth, the 
rest in gender 
violence)
Civil 
organization 
co￾investment86
416,100 MXN
(53% invested in 
youth at risk and the 
rest in CPTED)
6,651,764 MXN
(95% invested in at-risk
youth, the rest in 
sexual violence)
1,791,162 MXN
(88% invested in at-risk 
youth, the rest is a co￾investment from the 
anchor actor)
1,598,624 MXN
(97% was for at-risk 
youth, the rest a co￾investment from the 
anchor actor)
- 473,020 MXN
(70% is a co￾investment from 
the anchor actor, 
the rest was 
invested in at-risk
youth) 
Academia co￾investments
3,000 
(invested in CBT)
- - - 305,000 MXN
(100% in civic justice)
-
JPV donations 3,132,751.38 
MXN (49% invested 
in territorial 
strategies and 51% in 
focalized strategies, 
specifically CBT￾informed 
interventions, as part 
of the civic justice 
model)
6,451,630.16 MXN
(42% invested in 
focalized strategies for 
youth in conflict with 
the law, and 58% in 
territorial strategies)
5,864,589.04 MXN
(84% invested in 
focalized strategies, in 
particular CBT for youth 
in conflict with the law 
and adapted to be 
implemented in schools, 
and the rest in territorial 
strategies)87
5,716,784.04 
MXN
(65% invested in 
focalized strategies—
35% for the 
adaptation of a CBT￾informed 
intervention for civic 
justice, 17% for a 
CBT-informed 
intervention in 
schools, and 17% for 
an intervention for 
youth in conflict with 
the law. The rest was 
invested in territorial 
1,630,558.45 MXN
(43% invested in 
focalized strategies 
for civic justice, in 
particular a CBT 
intervention and an 
addiction program. 
The rest was invested 
in a territorial 
strategy)
4,999,580.74 
MXN (62% 
invested in 
focalized strategies 
for youth in 
conflict with the 
law, 20% for a 
focalized strategy 
of a CBT 
intervention in 
schools, and 18% 
in a territorial 
strategy)
85 Additional 1,690,866 MXN was leveraged in Mexico City 86 Additional 1,576,372MXN was leveraged in Mexico City 87 This number considers the subcontract with ProSociedad that was aimed at developing a CBT intervention in schools in collaboration with the government. 
116
Table 1. Comparative Municipalities Matrix
Feature / 
region
Chihuahua Ciudad Juárez Guadalajara
(metropolitan 
region)
Monterrey
(metropolitan 
region)
Morelia Tijuana
strategies related to 
new masculinities)
Local trust A local trust was established in 2010 for crime 
and violence prevention—FECHAC and 
FICOSEC.
In 2019 a fund was 
inaugurated by the 
anchor actor to finance 
crime-and-violence￾prevention initiatives in 
the region—with a 
communication strategy 
to attract more funds.
The United 
Businesspeople, a 
nonprofit in 
Escobedo, was 
inaugurated in 2019,
but the financial 
sustainability plan is 
still developing. 
Trust was established 
in 2019 but given the 
COVID-19 crisis 
funds are unavailable 
(the payroll tax—the 
source of funding—
was suspended).
Efforts to build a 
local trust haven’t 
been successful 
yet. 
117
ANNEX V. DONATIONS PROGRAM
Table 1. Donations Program Grants and Products
RFP Technical agenda Grants (region) Systematization/ 
Publication
001
2016-17
Evidence-bases models, including:
CBT, Parental skills development, 
alcohol and drug abuse, mentorship, 
restorative justice, social cohesion, new 
masculinities, social reinsertion, focused 
deterrence, procedural justice, gender￾based violence, environmental 
prevention policies.
This call invited programs with 
implementation timelines of maximum 
one year, with the opportunity to renew 
for two additional phases for a 
maximum, each of one year (2017 and 
2018)
Total granted amount:
MXN 11,310,226
CEDAT (Guadalajara)
Strengthen the intervention model based on 
mindfulness and CBT components aimed at 
reducing the incidence of juvenile delinquency. 
* This grant had a follow-up award to replicate 
and adapt the model for the adoption by other 
organizations, including a treatment option for 
the Judicial System in the State of Jalisco, with 
a total amount grated MXN 524,5050 (not 
included in the total amount in the right column).
Model 
systematization
Implementation 
manual
Impact
evaluation
ITESM (Morelia)
Feasibility study of evidence-based programs/
study on the ecosystem development based on 
results and implementation opportunities.
Diagnosis 
publication
TELAR DEL SARAPE (Guadalajara)
Strengthening of the social transformation 
model, based on creativity and neuro￾education as tools for social transformation
Model 
systematization 
ACCIONA (Morelos)
Strengthening of the violence prevention 
model based on mentorship provision aimed at 
finding alternative activities options according 
to youth’s needs and capabilities.
Model 
systematization 
GENTE DIVERSA (Baja California –
including Mexicali, Rosarito, Tecate, 
Tijuana, and Ensenada)
Develop guides for the application of the Safe 
Cites Model aimed at preventing and 
eradicating violence against women in urban 
areas with high crime incidence
Results 
evaluation
ODP (Escobedo)
Develop a curriculum and Implement CBT 
interventions for young people in situations of 
vulnerability of violence. Develop 
Model 
systematization
Implementation 
manual
IPAP (Morelia, Tijuana, Mexico City, 
Chihuahua, Guadalajara, Monterrey, 
Ciudad Juárez)
Implement the project “Best local prevention 
practices for gender based violence” to 
identify practices implemented at the local 
level.
Model 
systematization 
(not public)
118
Table 1. Donations Program Grants and Products
RFP Technical agenda Grants (region) Systematization/ 
Publication
SIMO (Chihuahua)
Support the Unit of Criminal Analysis develop 
their data management system to be used as 
an input for prevention policies.
-
SIMO (Tijuana)
Design, implement, and evaluate a CBT 
intervention targeted at youth in urban 
schools to prevent violence.
Model 
systematization
(not public)
Results 
evaluation
(not public)
REINTEGRA (Mexico City)
Strengthen the reintegration model for youth 
in conflict with the law under non-custodial 
measures
Model 
systematization
Implementation 
manual
RENACE (Monterrey)
Strengthen the model aimed at social 
reintegration of youth un conflict with the law 
through counseling and services, such as 
accompaniment with youth and their families 
during their trial and sentencing, as well 
subsequent follow-up activities.
Model 
systematization 
(identified as 
good practice)
RED DE COHESION (Chihuahua)
Strengthen the interventional model aimed at 
addressing problems of violence through the 
rescue and recovery pf public spaces, by 
introducing the YTT and activities to work 
with youth displaying high-risk behaviors-
-
002
2017
Models focused on social reinsertion 
through programs focused on generating 
employment opportunities for youth in 
secondary or tertiary prevention and in 
conflict with the law.
Total granted amount:
MXN 5,843,412
VICTORIA EMERGENTE (Mexico City)
To promote the economic and social inclusion 
of youth who are part of cultural groups by 
implementing an alternative pedagogical model 
that strengthen s the technical and
psychological capacities with a focus on 
violence prevention through socio￾entrepreneurship productive projects.
Model 
systematization 
(not public)
CASA (Ciudad Juárez)
Implement a model aimed at interrupt cycles 
linked to violence and acts of crime committed 
by at-risk youth, by developing resilient 
identities, options for educational and job 
reinsertions, and prevention capacities in 
communities.
Model 
systematization 
(not public)
SUPERA (Monterrey) Results 
evaluation
119
Table 1. Donations Program Grants and Products
RFP Technical agenda Grants (region) Systematization/ 
Publication
Implement the New Masculinity “Youth 
Transforming Youth” program aimed at 
reducing gender based violence.
(not public)
004
2017
Models focused on social reinsertion 
through programs focused on generating 
employment opportunities for youth in 
secondary or tertiary prevention and in 
conflict with the law, or in advanced 
processes of addiction treatment.
Total granted amount:
MXN 5,685,349
FUNDACION COMUNITARIA DE LA 
FRONTERA NORTE (Ciudad Juárez)
Implement the model “Desafío” aimed at 
engaging at-risk youths in productive and 
participative activities while strengthening 
protective factors against violence and 
delinquency, though psychological support, the 
development of interpersonal skills through 
sports, technical in-job training.
Model 
systematization
Results 
evaluation
(not public)
TENDA DI CRISTO (Ciudad Juárez)
Implement the model “Yo Soy Rediseño 
Social” aimed at reducing recidivism in youth 
in conflict with the law. Based on multi-actor 
collaboration processes seeking to increase 
social inclusion (socio-educational and 
employment). 
Results 
evaluation 
(not public)
PROSALUD (Monterrey)
Implement the model “Orale” which is an 
adoption of the International Youth 
Foundation’s Youth Employment Model, aimed 
at increasing employment and educational 
opportunities for at-risk youth, teaching soft 
skills. This model also added CBT components 
to enhance impact.
Results 
evaluation
(not public)
Total granted amount:
MXN 2,273,109
FUNDACION EDUCACION 
SUPERIOR EMPRESA (Chiapas and 
Tijuana)
Implement care models aimed at the social 
reintegration of youth in conflict with the law 
and or at-risk to create opportunities for labor 
inclusion.
-
006
2019
Development and implementation of a 
CBT model for youth between 18 and 
29 years old that are derived from the 
Civic Justice System in the cities of 
Chihuahua and Escobedo (one on one), 
with the objective of reducing their 
impulsivity and improve self-control.
Total granted amount:
MXN 3,298,431
LIBRES POR AMOR (Chihuahua)
Develop or adapt a model with CBT 
components to assist youth from 18 to 29 
years old derived from the Municipal Civic 
Court in Chihuahua.
In progress
SUPERA (Monterrey)
Develop or adapt a model with CBT 
components to assist youth from 18 to 29 
years old derived from the Municipal Civic 
Court in the Metropolitan Area of Monterrey.
In progress
007
2019-20
1) Model of addiction treatment 
2) CBT model
To work with youth between 18 and 29 
years derived from the Civic Justice 
System in the City of Morelia
Universidad Vasco de Quiroga - UVAQ 
(Morelia)
Develop or adapt a model with CBT 
components to assist youth from 18 to 40 
years old derived from the Municipal Civic 
Court in Morelia.
In progress
120
Table 1. Donations Program Grants and Products
RFP Technical agenda Grants (region) Systematization/ 
Publication
Total granted amount:
MXN 1,818,313
CIJ (Morelia)
Develop or adapt a model with addiction 
treatment to assist youth from 18 to 40 years 
old derived from the Municipal Civic Court in 
Morelia.
In progress
008
2019
Reintegration model, at the community 
level, of men and women at risk to 
generate violent behaviors, in conflict 
with the law, or substance consumption, 
in Tijuana
Total granted amount:
MXN 899,560
PROSALUD (Tijuana)
Implement a model for the reintegration of 
young people, though the development of a 
diagnosis of main risk factors, development of 
a canalization protocol, implement an 
incorporation program, and generation of 
strategic alliances to channel youth for other 
services.
Model 
systematization
009
2019
Care model to accompany the
process of comprehensive social 
reintegration of young people in conflict 
with the law on non-custodial sanctions 
in two municipalities of the Guadalajara 
Metropolitan Area, through the
accompaniment one on one mentoring 
to develop their educational, work / 
production environment,
personal and family.
Total granted amount:
MXN 1,690,945
CEDAT (Guadalajara)
Implement a program of accompanying young 
people in conflict with the law on non￾custodial sanctions.
See above on 
CEDAT’s row
010
2019-20
To give continuity to the care model 
aimed at the social reintegration of 
young people in conflict with the law and 
/ or at risk through programs to 
generate opportunities for job inclusion 
(Hope Truck) – in Tijuana.
Total granted amount:
MXN 1,328,407
TIJUANA INNOVADORA (Tijuana)
Give continuity to the care model aimed at the 
social reintegration of young people in conflict 
with the law and / or at risk through programs 
to generate opportunities for job inclusion 
(Hope Truck) – in Tijuana.
In progress
011
2020
For the elaboration of an Intervention 
Manual of the Model of the Supervision 
System of Adolescents in non-custodial 
measures the State of Chihuahua, as well 
as the Manual for the use of YTT tool as 
part of this model. 
Total granted amount:
MXN 800,323
TENDA DI CRISTO (Ciudad Juárez)
A follow up grant to “Yo Soy Rediseño Social” 
to meet the demand for channeling youth with 
non-custodial measures.
In progress
121
ANNEX VI. JPV ORGANIZED EVENTS AND STUDY TOURS
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
At-risk youth Events Mexico City: In 
collaboration with J￾PAL, JPV organized 
the workshop “The 
Use of Evidence in 
Preventing Youth 
Violence: Challenges 
and Opportunities in 
Latin America" in 
Mexico City. More 
than 100 attendees 
shared experiences 
and lessons learned 
from implementing 
CBT interventions in 
Mexico, El Salvador, 
Honduras, the 
United States, and 
Liberia (2017).
Mexico City: In collaboration with J￾PAL, JPV organized the seminar 
“Prevention of Youth Violence: 
Challenges and Opportunities in Mexico 
and Latin America” in Mexico City to 
present the results of What Works to 
Prevent Violence Among Youth? Speakers 
included professor Thomas Abt from 
Harvard University, Beatriz Magaloni
from Stanford University, and staff from 
the University of Chicago’s crime lab.
Ciudad Juárez: FICOSEC Juárez 
organized a seminar on program 
evaluation for interventions focused on 
violence prevention during the 
evaluation week for Latin America and 
the Caribbean. Relevant actors of other 
regions such as CEDAT, RENACE, and 
COLEF88 participated as speakers.
Tijuana: JPV and Tijuana Innovadora
organized the first International Summit 
on Citizen Security to discuss promising 
crime-and-violence-reduction policies 
such as CPTED, focused deterrence, 
88 The College of the Northern Border (COLEF for its acronym in Spanish) 
122
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
and civic justice. The summit gathered 
more than 150 participants.
Ciudad Juárez and Guadalajara: 
FICOSEC traveled to Guadalajara to 
learn about CBT interventions 
implemented by CEDAT and 
ProSociedad. This study tour led to the 
future transfer of CEDAT’s model.
Guadalajara and Ciudad 
Juárez: A group of actors 
from Guadalajara traveled to 
Ciudad Juárez to meet and 
learn from the Ciudad Juárez 
Social Reintegration 
Committee; this study led to 
the creation of a Social 
Reintegration Committee for 
Youth within the local system 
in 2019.
Civic justice (or 
related, including 
police reform)
Events Guadalajara: JPV, the Metropolitan 
Security Agency, and the University of 
Guadalajara organized the Civic Justice 
Homologated Model, Good Governance 
and Culture of Legality of the 
Guadalajara Metropolitan Area Forum.
Speakers included municipal authorities 
from Escobedo and Morelia to share 
their experiences in the implementation 
of the civic justice model; around 100 
people attended.
Morelia: JPV and the municipal 
government of Morelia organized the 
first International Security and Civic 
Justice Forum. More than 300 
participants attended the event to share 
results and challenges faced 
implementing the civic justice model.
Monterrey: JPV, Mesa 
Metrópoli and the municipal 
government of Escobedo 
organized the International 
Forum for Violence Prevention 
and Citizen Security in 
Escobedo. Speakers included 
Julius Lang, Senior Advisor for 
Training and Technical 
Assistance at the Center 
for Court Innovation; 
David M. Kennedy, Director of 
the National Network for Safe 
Communities at John Jay 
College of Criminal Justice in 
New York; and government 
authorities. Around 400 
participants attended the event 
to discuss evidence-based 
violence prevention strategies
Guadalajara: JPV, the 
municipal government of 
Morelia, and the 
municipal government of 
Tonalá organized the
International Forum of 
Problem-Oriented 
Policing and Civic Justice 
(2020). During the event, 
representatives of local 
governments exchanged 
experiences to 
strengthen the 
implementation of the 
Civic Justice 
Homologated Model.
Monterrey: JPV and 
Mesa Metrópoli 
organized the forum 
123
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
in Mexico and the world, such 
as focused deterrence, civic 
justice, and proximity police.
Promising Practices in 
Security and Justice: 
From a Local 
Perspective. Speakers 
included representatives 
of local governments 
such as Nuevo León, San 
Pedro, and Allende, 
among others.
Study 
tours
Ciudad Juárez and Morelia: A 
delegation of public servants from the 
Municipal Government of Juárez 
traveled to Morelia to participate in 
the Civic Justice and Security Forum. 
They took advantage of the opportunity
to learn more about the implementation 
of the Civic Justice Homologated Model.
Monterrey: Representatives of the 
Municipal Government of Escobedo 
traveled to New York to learn from 
the Courts Innovation Center about the 
different models that exist on 
community justice courts to adopt 
practices in the Mexican context.
Gender strategy Events Chihuahua: The Social
Prevention of Violence and 
Crime Council of Chihuahua
(COPREV being its Spanish 
acronym), with support from 
FICOSEC and USAID 
organized the Forum on Best 
124
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
Practices and the Prevention of 
Domestic Violence.
Ciudad Juárez: JPV, the 
Municipal Women’s Institute, 
and the Autonomous 
University of Ciudad Juárez 
organized the Forum on 
Strategies to Reduce Gender￾Based Violence to share 
experiences and challenges 
with local actors. 
Study 
tours
Ciudad Juárez: Representatives of 
Promotora Social México, the 
Corporate Social Responsibility branch 
of Gentera Bank, went on a visit to 
Ciudad Juárez to learn about the work 
of FICOSEC, Casa Amiga, La Tenda de 
Cristo, and Fundación Paso del Norte 
para la Salud y Bienestar. After this visit, 
Promotora Social México agreed to 
finance two of these organizations in 
their efforts to eradicate sexual violence
among children. Fundación Paso del 
Norte para la Salud y Bienestar 
committed to invest an additional 50% 
of the funding provided by the bank.
Monterrey: During the 
International Forum for 
Violence Prevention and 
Citizen Security in Escobedo, a 
study tour was conducted to 
the local security center (C4) 
and to a care center that treats 
women who are victims of 
violence.
Environmental design Events Guadalajara: JPV, the Metropolitan 
Security Agency, and the University of 
Guadalajara organized the Urban Design 
and Violence Prevention Forum: 
Experiences in Mexico. Speakers 
included representatives from 
interventions in Chihuahua, Juárez, and 
125
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
Tijuana. More than120 participants 
attended the event.
Institutional￾strengthening 
strategy
Events Mexico City: JPV and the 
Monterrey Institute 
of Technology organized an 
event in Mexico City to launch 
the virtual learning community.
Mexico City: JPV organized a 
meeting in Mexico City for 
representatives from local 
organizations based in Morelia, 
Ciudad Juárez, Jalisco, 
Monterrey, Chihuahua, and 
Tijuana to exchange best 
practices on violence 
prevention.
Study 
tours
Ciudad Juárez:
Juárez Mágico was a 
study trip organized 
by JPV and FICOSEC 
that brought 
businessmen and 
representatives of 
CSOs and business 
chambers from 
Tijuana, Nuevo 
León, Irapuato, and 
Morelia to Juárez to 
learn about 
FICOSEC’s work. It 
sought to increase 
collaborations in 
Morelia and Chihuahua: Members of 
Morelia’s municipal security commission 
traveled to Chihuahua to learn from 
FICOSEC’s crime observatory to foster 
knowledge generation on public security 
and civic justice.
Morelia and Chihuahua: A
group of entrepreneurs 
traveled to Chihuahua to visit 
FICOSEC and FECHAC to 
learn from its model to fund 
prevention strategies and 
replicate it.
126
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
crime and violence 
prevention, 
strengthen LPS 
capacities, and 
enhance private 
sector participation 
in this agenda. 
Participants then 
visited the project 
Juárez Mágico, an 
environmental design 
prevention project.
Several other 
exchanges followed 
this tour:
- Toks89 visited 
FECHAC to share 
work strategies
- Pro-Irapuato visited 
Tijuana Innovadora 
to learn from its 
organizational 
structure.
- CINLAAC90 funded 
by FEMSA,91 visited 
FICOSEC in
Chihuahua to learn 
from its model.
89 Restaurant chain 90 Civic Council of the Institutions of Nuevo León (CINLAAC for its acronym in Spanish) 91 Fomento Económico Mexicano (a Mexican multinational beverage and retail company headquarteed in Monterrey, Mexico) 
127
Table 6. Events and Study Tours, by Transversal Strategy
Strategy Type 
of 
activity
2017 2018 2019 2020
International sharing 
of experiences 
Study 
tours
Monterrey: Public servants from 
Escobedo’s municipal government 
presented their experience 
implementing the Civic Justice 
Homologated Model in the Public Safety 
and Community Courts Conference 
organized by the Courts Innovation 
Center in Alabama.
Morelia: Representatives of the federal 
government of Guatemala traveled to 
Morelia to identify and learn from 
promising interventions with the 
intention to replicate them in the
Guatemalan context, one of them being 
civic justice.
Monterrey: Representatives 
of the federal government of 
Jamaica and USAID’s staff in 
Jamaica visited Guadalupe to 
learn from Supera’s Youth 
Transforming Youth model.
128
ANNEX VII. JPV’S RESOURCES DEVOTED TO TRANSFER AND SCALE-UP AND RESULTS
Table 1. JPV’s Resources Devoted To Transfer and Scale-Up and Results
Best practices 
in violence and 
crime 
prevention 
Knowledge development Implementation of Solutions & 
Improve the practice (policy 
influence)
Improve the practice 
(Publications to support 
adaptation/transfer)
Solutions for at￾risk youth
Main result:
JPV invested
significant 
effort in 
developing, 
systematizing, 
scaling, and 
transferring
CBT-informed 
interventions. 
CBT-informed 
interventions
are now being 
implemented 
by several 
organizations 
across the 
regions.
JPV financed five different CBT 
interventions (SIMO,
92 ODP93, Reintegra, 
CEDAT, and ProSociedad). Two received 
special assistance for the scale-up and 
adaptation to different settings: CEDAT’s 
Mente Libre and ProSociedad’s ROLE.
- CEDAT developed the Mente Libre 
model based on CBT and mindfulness 
components. It is focused on reducing 
young felons’ propensity to 
committing crimes.
- ProSociedad developed the ROLE 
model, focused on reducing young 
men’s exposure to violence in urban 
contexts through tools that change 
cognitive and behavioral processes, 
along with mechanisms that generate 
more functional, prosocial, and 
healthy lifestyles.
JPV established a collaboration with J-PAL 
to facilitate knowledge exchange with 
international organizations implementing 
CBT and academics to learn from existing 
experiences and evidence. Through this 
collaboration, local CSOs also received 
CEDAT transferred its Mente Libre
model to Jalisco’s internment 
centers and is currently working to 
transfer the model to organizations 
in Chihuahua and Ciudad Juárez, 
and 10 additional municipalities in 
Jalisco. 
- In particular, Libres por Amor 
with technical assistance from 
CEDAT developed an 
intervention model with CBT 
and mindfulness components 
to assist at-risk youth derived 
from the municipal civic court 
in Chihuahua.
- CEDAT and FICOSEC
launched a call for proposals 
invited organizations to 
participate in a program aimed 
at transferring the model to 
interested organizations. The 
programs include in-person
training and online 
accompaniment. 
Mente libre
Implementation materials:
Instructor's Manual: Mente Libre 
Protocol
Youth and violence in Mexico: 
Dialogue between neurosciences, 
civil society, and social sciences
Evaluation:
Has an internal experimental 
evaluation (caveat: sample size is very 
small).
ROLE:
Iimplementation materials
ROLE Implementation Manual
M&E ROLE Manual
Evaluation:
Has an external process and results 
(quasiexperimental) evaluation of the 
model implemented in schools, but 
92 Intelligence Systems in Markets and Opinion (SIMO being its acronym in Spanish) 93 Observatory of Regional Development and Social Promotion (ODP being its acronym in Spanish)
129
technical assistance for their program 
design. 
CEDAT and ProDociedad received 
further support to systematize and 
rigorously evaluate their models and 
develop manuals for their adoption by 
other organizations. ProDociedad’s model 
was evaluated by an external evaluator, 
funded by JPV. 
Once the models were operating, JPV
organized events and study tours to 
facilitate knowledge exchange between 
regions.
ProSociedad’s ROLE model was
transferred to high schools in 
Jalisco. The government mandated 
a statewide institutionalization of 
the model in coordination with the 
Ministry of Education, and 
ProSociedad supported the process 
and results evaluation of the 
adaptation process.
- In particular, ProDociedad 
transferred its model to 
Supera in Monterrey to assist 
youth at risk derived from the 
municipal civic court in 
Chihuahua.
- ProSociedad is also 
collaborating with UVAQ in 
Morelia to transfer the ROLE 
model to its CAS in Morelia, 
with the objective of 
strengthening the civic justice 
system in the city by increasing 
its capacity to receive and 
treat offenders.
not for the models being replicated 
for youth in conflict with the law.
Civic justice
Main result:
JPV promoted 
the Civic 
Justice 
Homologated 
Model based on 
the innovative 
approach of 
public officials 
in Morelia to 
deescalate 
violence and 
conflict and 
The Civic Justice Homologated Model was 
promoted by JPV based on the innovative 
approach of public officials in Morelia to 
deescalate violence and conflict and offer 
an alternative for everyday conflict 
resolution. In addition, JPV incorporated 
the PoP approach to strengthen the role 
of the police within the framework of civic
justice.
JPV worked with the Mexican 
government, through the National
Commission of Public Security, to design 
the Civic Justice Homologated Model,
which was approved in 2017. From this 
Because of high demand, JPV 
started developing civic justice 
model implementation manuals for 
judges, mediators, public security 
officers and other local judicial 
workers. In addition to supporting 
the materialization of the idea into 
a replicable and standardized 
intervention model, JPV worked 
jointly with the federal government 
to lead working groups with 27 
municipalities. Furthermore, JPV 
organized seminars and exchange 
opportunities with key 
international actors in community 
Implementation materials:
Guide I. Diagnosis
Guide II. Regulatory adequacy
Guide III. Public hearings
Guide IV. Convivencia Cotidiana
Model: Police Oriented to Problem 
Solving for Civic Justice
Implementation Guide: Problem￾Oriented Police for Civic Justice: 
130
offer an 
alternative for 
everyday 
conflict 
resolution. JPV 
worked to 
developed and 
systematize the 
system, also 
incorporating 
the PoP
approach to 
strengthen the 
role of the 
police within 
the framework.
The Civic 
Justice 
Homologated 
Model is now 
being scaled up 
to more than 
15 
municipalities 
in the country 
with additional 
demand for 
replication in 
others. 
initiative, the Mexican government
determined that federal 
funds will be allocated to the FORTASEG 
to implement the Civic Justice 
Homologated Model.
In addition to the work at the national 
level, JPV worked with and monitored the 
progress of, the municipal governments of 
Morelia and Escobedo to implement pilots 
of this model. JPV’s technical assistance, 
which was based on the recruitment of 
specialized consultancies, culminated in 
the development of an institutionalized 
mechanism, which then provided the 
knowledge basis to transfer the model to 
other areas. 
justice, developed exhaustive 
guides that detail the operational 
requirements to implement the 
model, and developed focalization 
tools to assess the risk level of
offenders.
Besides Morelia and Escobedo, JPV 
provided technical assistance and 
recruited specialized consultant 
services to provide support to 
Chihuahua, Ciudad Juárez, the 
Guadalajara metropolitan area, the 
Monterrey metropolitan area, and 
Los Cabos to design, implement, 
and systematize the intervention 
model. 
In 2018, after the change in the 
federal administration, the Mexican 
government included the Civic 
Justice Homologated Model and 
CBT in its list of priorities for the 
allocation of funds of FORTASEG 
and the Public Security 
Contribution Fund (FASP94 being its 
acronym in Spanish).
In July 2019, the Mexican 
government launched the new 
National Police and Civic Justice 
Model, which incorporates in its 
entirety the Civic Justice 
Homologated Model and the PoP
model supported by JPV. By 2020, 
36 municipalities had allocated 
FORTASEG resources to the Civic 
Justice Homologated Model.
Adjustments to the Regulatory 
Framework
Module 1. Participant's Guide: Police 
Oriented to Conflict Resolution in 
Communities
Module 1. Instructor's Guide: Police 
Oriented to Conflict Resolution in 
Communities
Module 2. Participant's Guide: Police 
Oriented to Problems Solution with
Mediation Capacity
Module 2. Instructor Guide: 
Participant's Guide: Police Oriented 
to Problems Solution with Mediation 
Capacity
Module 3. Participant's Guide: PoP 
model tools
94 Fondo de Aportaciones para la Seguridad Pública in Spanish
131
In late 2019 and 2020, the Mexican 
government and JPV have agreed to 
provide technical assistance to ten 
more municipalities on the Civic 
Justice Homologated Model: 
Acapulco, Cancun, Ciudad Juárez, 
Hermosillo, La Paz, Los Cabos, 
Manzanillo, Playa del Carmen, 
Puerto Vallarta, and Tijuana.
Environmental 
prevention 
strategy
JPV provided technical assistance and 
support to CSOs and municipal 
authorities to implement interventions 
under the CPTED methodology in 
Chihuahua, Guadalupe, Mexicali, Tijuana, 
and Zapopan with a focus on at-risk youth 
and GBV. 
JPV’s technical assistance consisted 
primarily of grants to implementing 
organizations, the systematization of 
promising interventions following the 
CPTED methodology, and the training and 
CPTED certification by specialized service 
providers of local actors of Tijuana, 
Ciudad Juárez, Chihuahua, Monterrey, and 
Guadalajara. This certification will enable 
participants to apply the CPTED approach 
in their LPSs to implement sustainable 
crime-and-violence-prevention strategies.
As a result of the evaluation of the 
CPTED model My neighborhood,
My House, Chihuahua’s municipal 
government decided to expand and 
replicate the model in four new 
neighborhoods. In order to 
replicate the model, public officials
from Chihuahua participated in a 
series of practical and theoretical 
trainings on the CPTED model by 
JPV. The replication of this 
intervention model has resulted in 
each neighborhood applying a 
participatory diagnostic, 
establishing a baseline, developing
an annual work plan with indicators 
and goals, and implementing actions 
to enhance trust with the local 
stakeholders.
In Tijuana, BBVA-Bancomer, in 
collaboration with Tijuana 
Innovadora (the anchor actor)
developed the model Mejora 
Mutualista, which is informed in 
CPTED principles. Also, Gente 
Diversa developed the model 
Secure Cities for Women for 
Implementation materials:
Implementation guide of the CPTED 
strategy from a prevention 
perspective for GBV, at-risk youth, 
and the community asset 
construction model
Implementation manual: Mejora 
Mutualista—BBVA
Situational prevention: Safe Corridor 
for Ciudad Juárez women
Evaluation:
External process/ pre- postevaluation 
(in collaboration with FICOSEC, the
anchor actor)
132
several cities in Baja California, 
which is informed by CPTED 
principles. 
Also
, Fundación Tú+Yo has 
developed the model Abordaje Comunitario, informed by CPTED 
principles. It is implemented in 
Tijuana, Mexicali, Chihuahua, and 
Ciudad Ju
árez 
In Zapopan
, a public
-private 
partnership emerged between the 
Zapopan’s Public Space Authori
ty 
and CEMEX—a multinational
company that specializes in the 
production and sale of concrete 
and similar materials
—to 
implement the workshop “I build
,” 
which develops construction skills 
among vulnerable populations of 
the Miramar neighborhood for the 
urba
n design of safer communities. 
Given the success of the project, 
CEMEX financed a second 
implementation, valued at 
40,000.00 MXN (about 2
,000
USD).
Gender strategy
Main result:
JPV developed 
the Care Model 
for Child and 
Adolescent 
Victims of
Family 
Violence
Through the recruitment of expert 
consultants and in collaboration with women’s justice centers, JPV supported 
the piloti
ng and systematization of the
care 
model in Ciudad Juárez and Pachuca, 
which includes CBT components, group 
therapy methods, standardized tools for 
monitoring service users
, and program 
evaluation methods.
As a result of the piloting of the 
model, the 
care 
model was 
transferred to institutions in three 
additional local systems: Ciudad 
Ju
árez, Monterrey, and Gu
adalajara.
JPV organized trainings for 
psychologists and public officials 
with local allies that can further 
increase the reach and impact of 
the strategy. A total of 16 
women’s 
center
s were trained on the model.
133
The program also launched Chihuahua’s 
Sexual Violence Prevention Strategy, 
which got the commitment of the state of 
Chihuahua. In alliance with the National 
Pedagogical University of the State of 
Chihuahua, JPV developed the Diploma on 
Comprehensive Care for Sexual Abuse 
during Childhood and Adolescence and 
provided specialized training to 200 public 
servants. It leveraged 6,000,000 MXN 
from local investment to implement 
related strategies.
134
ANNEX VIII. TARGETING ANALYSIS
1. HOMICIDE TOTALS BY SOURCE
Table 1: Comparison of Intentional Homicide Data from INEGI, SESNSP, and Local 
Police Departments
Year / City Chihuahua Escobedo Guadalupe
I P S I P S I P S
2015 253 188 142 21 29 24 24 32 27
2016 409 300 275 22 30 23 45 73 55
2017 535 389 363 59 78 62 52 71 62
2018 479 313 282 42 56 53 66 78 77
2019 160 332 44 62 78 83
1676 1350 1394 144 237 224 187 332 304
Year / City Monterrey Ciudad Juárez Tijuana
2015 181 161 130 440 383 269 644 612 612
2016 246 1014 209 680 549 470 857 871 871
2017 193 156 145 805 737 642 1669 1618 1618
2018 323 283 236 1441 1139 1032 2302 2252 2253
2019 178 239 644 1293 2012 2013
943 1792 959 3366 3452 3706 5472 7365 7367
Source KEY: I = Mexico’s National Institute for Statistics and Geography (INEGI, for its acronym in 
Spanish), S= Mexico’s Secretary General of National Public Security (SESNSP, for its acronym in 
Spanish), INEGI, P = Local Police.
1I. COVERAGE BY MUNICIPALITY 
II.1 CHIHUAHUA
Table 2: Most Common Neighborhoods Registered in 
Chihuahua
11 de febrero 18
Riveras de Sacramento 16
Villa Juárez 15
Punta Oriente 14
Chihuahua 2000 13
Popular 4
135
Table 3: Intentional Homicide, Top 10 Neighborhoods in Chihuahua
Neighborhood / Year 2014 2015 2016 2017 2018 2019 Grand 
Total
Centro 10 13 9 16 19 67
Sin Datos Específicos 3 1 8 10 20 23 65
Vía Pública 1 11 21 33
Obrera 5 6 11 6 2 1 31
Granjas del Valle1 8 5 8 7 3 31
Revolución 6 3 6 6 2 3 26
Cerro de la Cruz 3 3 8 5 3 1 23
Villa Juárez 2 4 3 7 2 2 20
Aeropuerto 3 2 1 6 2 5 19
20 Aniversario 1 1 4 2 6 4 18
Total Homicides (all 
neighborhoods)
209 188 399 389 313 160 1,350
II.2 CIUDAD JUÁREZ
Table 4. Most Common Beneficiaries’ 
Neighborhoods Registered in Ciudad 
Juárez
Díaz Ordaz 117
Adolfo López Mateos 81
La Reforma 67
Cersai 3 56
16 de Septiembre 52
Riveras del Bravo 22
Guadalajara Izquierda 22
Plutarco Elías Calles Alta 17
Table 5. Intentional Homicides in Ciudad Juárez by Neighborhood
Neighborhood / Year 2014 2015 2016 2017 2018 2019 Grand Total
Margaritas 16 6 7 27 39 9 104
Centro 15 2 4 7 22 15 65
Salvarcar 8 6 8 9 16 4 51
Riveras De
l Bravo
9 1 4 6 22 9 51
Infonavit Casas Grandes 5 7 4 8 18 5 47
Paseo De Las Torres 2 1 4 10 15 7 39
Chaveña 6 1 3 3 14 11 38
Zona Pronaf 18 2 2 3 10 2 37
La Cuesta 4 3 7 1 16 6 37
Aztecas 6 2 10 11 6 35
Total Homicides (all 
neighborhoods)
520 383 549 747 1,139 644 3,452
136
II.3 ESCOBEDO
Table 6: Most Common Neighborhoods Registered in 
General Escobedo
N/A 62
Alianza Real 17
Table 7: Intentional Homicides in General Escobedo by Neighborhood
Neighborhood / Year 2015 2016 2017 2018 2019 Grand Total
Villas de San Francisco 3 2 7 6 1 19
Andrés Caballero 1 2 7 4 14
Carretera 1 9 1 2 13
Alianza Real 6 2 2 10
Nueva Esperanza 1 2 2 2 7
Praderas de San Francisco 1 3 3 7
Nuevo Escobedo 2 4 6
La Unidad 2 1 2 1 6
Paso Cucharas Progreso 1 1 2 4
Total Homicides (all 
neighborhoods)
29 30 78 56 44 237
II.4 GUADALUPE
Table 8: Most Common Neighborhoods Registered in 
Guadalupe
Valle del Sol 119
Independientes 55
(blank) 32
Paseo de Guadalupe 31
Los Independientes 22
Santa Cruz 14
Jardines de Xochomilco 14
Table 9: Intentional Homicides in Guadalupe by Neighborhood
Neighborhood / Year 2015 2016 2017 2018 2019 Grand 
Total
Valle Soledo 1 5 3 5 14
Nuevo San Rafael FOM 7 1 1 6 8
Cañada Blanca 2 3 1 6
Centro de Guadalupe 2 2 2 6
Nuevo México 1 4 5
2 de Mayo FOM 20 3 2 5
Paseo de Guadalupe 1 1 2 1 5
Valle del Sol 1 2 1 1 5
Ciudad CROC 1 3 1 5
Total Homicides (all 
neighborhoods)
32 73 71 78 78 332
Santa Cruz ranks #18 on the list, not shown here.*
137
II.5 TIJUANA
Table 10. Most Common Beneficiaries’ 
Neighborhoods Registered in Tijuana
N/A 184
La Mesa 23
Los Álamos 23
Col. Luis Echeverria 22
Cañadas del Florido 19
Ejido Ojo de Agua 17
Camino Verde 14
Santa Fe 14
Table 11. Intentional Homicides in Tijuana by Neighborhood
Neighborhood / Year 2014 2015 2016 2017 2018 2019 Grand 
Total
Camino Verde 12 11 27 75 65 42 232
Zona Norte 7 26 43 49 43 41 209
Zona Centro 10 19 11 31 33 31 135
Sanchez Taboada 10 14 11 17 22 42 116
Tres De Octubre 3 4 6 12 38 36 99
Nueva Tijuana 5 7 7 19 29 18 85
Ejido Francisco Villa 7 3 8 16 25 21 80
El Florido Primera Y
Segunda Seccion
1 3 5 25 29 15 78
Mariano Matamoros 
Centro
6 7 10 14 21 15 73
El Pipila 2 4 4 12 25 25 72
Altiplano 5 3 5 18 15 10 56
Total Homicides (all 
neighborhoods)
462 612 871 1,618 2,252 2,012 7,365
138
ANNEX IX. PRELIMINARY STATISTICAL ANALYSIS
Figure 1. Homicide Rates by Municipality per 100,000 residents
2010-2018 (INEGI)
Figure 2. Perception of Population 18 Years and Older that Feels Insecure (ENSU)
0
50
100
150
200
250
300
2010 2011 2012 2013 2014 2015 2016 2017 2018
Chihuahua
Escobedo
Guadalajara
Guadalupe
Juárez
Monterrey
Morelia
Tijuana
Tonalá
Zapopan
Nacional
40
50
60
70
80
90
100
2016m3
2016m6
2016m9
2016m12
2017m3
2017m6
2017m9
2017m12
2018m3
2018m6
2018m9
2018m12
2019m3
2019m6
2019m9
2019m12
2020m3
Tijuana Chihuahua Juarez Tonala Zapopan
Morelia Escobedo Guadalupe Monterrey Mexico
139
Figure 3. Percentage of Population 18 Years and Older that Has Witnessed Alcohol 
Consumption in Public Spaces (ENSU)
Figure 4. Percentage of Population 18 Years and Older that Has Witnessed Drug Sells 
or Consumption (ENSU)
30
35
40
45
50
55
60
65
70
75
80
2016m3
2016m6
2016m9
2016m12
2017m3
2017m6
2017m9
2017m12
2018m3
2018m6
2018m9
2018m12
2019m3
2019m6
2019m9
2019m12
2020m3
Tijuana Chihuahua Juarez Tonala Zapopan
Morelia Escobedo Guadalupe Monterrey Mexico
0
10
20
30
40
50
60
70
80
2016m3
2016m6
2016m9
2016m12
2017m3
2017m6
2017m9
2017m12
2018m3
2018m6
2018m9
2018m12
2019m3
2019m6
2019m9
2019m12
2020m3
Tijuana Chihuahua Juarez Tonala Zapopan
Morelia Escobedo Guadalupe Monterrey Mexico
140
Figure 5. Percentage of Population 18 Years and Older that Has Witnessed the 
Existence of Street Gangs (ENSU)
Figure 6. Population 18 Years and Older that Has Engaged in Any Conflict (ENSU)
0
10
20
30
40
50
60
2016m3
2016m6
2016m9
2016m12
2017m3
2017m6
2017m9
2017m12
2018m3
2018m6
2018m9
2018m12
2019m3
2019m6
2019m9
2019m12
2020m3
Tijuana Chihuahua Juarez Tonala Zapopan
Morelia Escobedo Guadalupe Monterrey Mexico
0%
10%
20%
30%
40%
50%
60%
70%
80%
mar-16
jun-16
sep-16
dic-16
mar-17
jun-17
sep-17
dic-17
mar-18
jun-18
sep-18
dic-18
mar-19
jun-19
sep-19
dic-19
mar-20
Total	general
Tijuana Chihuahua Juárez Tonala Zapopan
Morelia Escobedo Guadalupe Monterrey Mexico