USAID/Kenya and East Africa (USAID/KEA) Kenya NiWajibu Wetu (NIWETU) Program Final Performance Evaluation September 2022 This document was produced at the request of the United States Agency for International Development. It was prepared independently by ME&A, Inc. FINAL PERFORMANCE EVALUATION USAID/Kenya and East Africa (USAID/KEA) Kenya NiWajibu Wetu (NIWETU) Program This publication was produced at the request of the United States Agency for International Development Developed under Contract Number: 72062320D00005/72061522F00002 It was prepared independently by: Sarah Wood, Team Leader Christine Kamau, Technical Specialist Suleiman Abdullahi, Technical Specialist Diana Ndung’u, Researcher Aisha Abdullahi, Researcher Contractor: ME&A, Inc. 4350 East-West Highway, Suite 210 Bethesda, MD 20814 Tel: 240-762-6296 www.MEandAHQ.com DISCLAIMER The author’s views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development or the United States Government. ACKNOWLEDGMENTS The NIWETU Evaluation Team (ET) would like to express appreciation to DAI, USAID/KEA, and John Langlois for assisting in the organization and stakeholder mapping of this evaluation. We are particularly grateful for the guidance and contributions given so freely to improve the utilization and value of the evaluation. We are grateful to the Wasafiri staff for taking the time to give feedback on the project design, partnership strategies, and monitoring methodologies of NIWETU to help the ET better understand NIWETU’s implementation context. We are grateful to the staff of the National Counter Terrorism Center (NCTC) and the Director and leadership staff of the Kenya School of Government, Security Management Institute (KSG/SMI) for their candor, wisdom, and insights on the past and current context of Countering Violent Extremism (CVE) in Kenya. We are especially appreciative of all the stakeholders and individuals who took the time to meet with us during a global pandemic and were so generous with their thoughts and recommendations—contributing to reducing and mitigating the risks of Violent Extremism (VE) in Kenya. CONTENTS EXECUTIVE SUMMARY................................................................................................................................................... i 1.0 EVALUATION PURPOSE AND QUESTIONS................................................................................................ 1 1.1 EVALUATION PURPOSE .............................................................................................. 1 1.2 EVALUATION AUDIENCE ............................................................................................. 1 1.3 EVALUATION QUESTIONS........................................................................................... 1 2.0 BACKGROUND ..................................................................................................................................................... 2 3.0 EVALUATION METHODS AND LIMITATIONS........................................................................................... 4 3.1 EVALUATION METHODOLOGY.................................................................................... 4 3.2 DATA COLLECTION...................................................................................................... 4 3.3.1 Data Collection ....................................................................................................... 5 3.3.2 Data Tools .............................................................................................................. 5 3.3.3 Data Analysis and Reporting................................................................................... 7 3.3.4 Evaluation Limitations and Mitigations .................................................................... 7 4.0 FINDINGS, CONCLUSIONS, AND RECOMMENDATIONS.................................................................... 8 4.1 FINDINGS...................................................................................................................... 8 4.1.1 Evaluation Question 1............................................................................................. 8 4.1.2 Evaluation Question 2............................................................................................10 4.1.3 Evaluation Question 3............................................................................................13 4.1.4 Evaluation Question 4............................................................................................14 4.1.5 Findings Specific to County Interventions ..............................................................16 4.2 CONCLUSIONS............................................................................................................19 4.2.1 Evaluation Question 1: P/CVE Design and Theory of Change Application.............19 4.2.2 Evaluation Question 2: P/CVE Interventions at the Hyper-Local Level...................22 4.2.3 Evaluation Question 3: Partnerships, Resilience, and J2SR ..................................24 4.2.4 Evaluation Question 4: P/CVE Targeting, Inclusion, and Marginalization...............26 4.3 RECOMMENDATIONS .................................................................................................27 4.3.1 Recommendations for P/CVE Design (Relevant to EQ1).......................................27 4.3.2 Recommendations for Interventions and Learning (Relevant to EQ2)....................27 4.3.3 Recommendations for Resilience and J2SR (Relevant to EQ3).............................28 4.3.4 Recommendations for Future P/CVE Targeting (Relevant to EQ4)........................28 4.3.4 Recommendations for Future P/CVE Programs.....................................................29 4.3.5 Improving CVE Programming in Development.......................................................29 ANNEXES ......................................................................................................................................................................... 32 ANNEX 1: STATEMENT OF WORK......................................................................................33 ANNEX 2: TOOLS.................................................................................................................41 ANNEX 3: STAKEHOLDERS CONTACTED .........................................................................46 ANNEX 4: DOCUMENT REVIEW..........................................................................................51 ANNEX 5: FIELD QUOTES...................................................................................................53 LIST OF ACRONYMS Acronym Definitions AMEP Activity Monitoring and Evaluation Plan AS Al-Shabaab 4C Champions for Change CAP Complexity Aware Planning CBO Community-Based Organization CDCS Country Development Cooperation Strategy CEF Community Engagement Forums COR Contracting Officer’s Representative COVID-19 Coronavirus Disease 2019 CSO Civil Society Organization CVE Countering Violent Extremism DAI Development Alternatives Incorporated DGC Democracy, Governance, and Conflict DDL Development Data Library DDR Disarmament, Demobilization, and Reintegration DEC Development Experience Clearinghouse DO Development Objective EoI Expression of Interest EQ Evaluation Question ET Evaluation Team FGD Focus Group Discussion GoK Government of Kenya HO Home Office IP Implementing Partner IR Intermediate Result J2SR Journey to Self-Reliance KEA Kenya and East Africa KII Key Informant Interview KSG Kenya School of Government M&E Monitoring and Evaluation Acronym Definitions ME&A ME&A, Inc. MEL Monitoring, Evaluation, and Learning MF Malaika Foundation MoI Ministry of the Interior and Coordination MOU Memorandum of Understanding NAPAD Nomadic Assistance for Peace and Development NCTC National Counter Terrorism Center NGO Non-Governmental Organization NIWETU Kenya NiWajibu Wetu Activity NSCVE National Strategy to Counter Violent Extremism OTI Office of Transition Initiatives PPE Personal Protective Equipment PPP Public Private Partnership PVE Preventing Violent Extremism RCAP Rapid County Action Plan SMI Security Management Institute SOW Statement of Work SPA Strategic Planning and Analysis TO Task Order ToC Theory of Change ToT Training of Trainers TL Team Leader USAID United States Agency for International Development USIU-A United States International University-Africa USG United States Government VE Violent Extremism VEO Violent Extremist Organization WFP World Food Programme i EXECUTIVE SUMMARY EVALUATION PURPOSE The purpose of this evaluation is to measure the effectiveness of the United States Agency for International Development/Kenya and East Africa (USAID/KEA)’s Kenya NiWajibu Wetu (NIWETU) Activity in achieving its key objectives: (1) Community mobilization to address Violent Extremism (VE) enhanced, and (2) Government responsiveness to VE improved. NIWETU, a four-year (2016–2020) activity, was designed to improve the capabilities of communities, subgrantees, and the Government of Kenya (GoK) to identify and respond to VE threats at the national and county levels with a focus on five core counties: Garissa, Isiolo, Nairobi, Mandera, and Wajir. EVALUATION AUDIENCE The main audience for this evaluation is USAID/KEA staff, who will use the findings to inform ongoing interventions designed to mitigate conflict and the design of new programs. ME&A, Inc. (ME&A), will also share evaluation results with a secondary audience—including USAID/Washington, USAID/East Africa, and other key stakeholders to include GoK ministries, departments, and agencies (in particular, the National Counter Terrorism Center [NCTC], Ministry of Interior, Coordination of National Government, and select county governments) and development partners. EVALUATION QUESTIONS USAID/KEA designed four main Evaluation Questions (EQs) to assess whether NIWETU had achieved its main objectives. The questions were: 1. How effective was the Theory of Change (ToC) in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved?’ 2. Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to violent extremism under Objective 1? How can these interventions be improved? 3. To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering violent extremism and advance its Journey to Self-Reliance (J2SR)? What do key counterparts perceive to be the positive and negative aspects of the partnership? 4. How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to violent extremism? How can these interventions be improved? EVALUATION DESIGN AND METHODOLOGY The ET employed a purposive sampling approach to ensure the inclusion of diverse beneficiary and stakeholder groups. The evaluation methodology ensured inclusive stakeholder mapping and sampling that was representative of the approach and partners of NIWETU in the five target counties,1 with key stakeholder groups including national and county-level government counterparts, civil society, and local partner organizations (Civil Society Organization [CSO] grantees and partners such as the Kenya School of Government [KSG] and NCTC, USAID/KEA, and DAI leadership). The ET conducted 80 Key Informant Interviews (KIIs), both virtually and in-person, as well as five Focus Group Discussions (FGDs) that comprised beneficiary populations of NIWETU CSO grants. The ET ensured that marginalized groups such as women, youth, minorities, and other vulnerable groups were represented. Of the KIIs conducted 26 percent were women, 74 percent were men, and 38 percent were youth. Selection of participants for 1 Considering the similarity of communities and the security situation in Garissa, Mandera, and Wajir counties, the ET plans to sample Garissa and Wajir counties, and, if needed, will cover Mandera through remote interviews with key beneficiaries and stakeholders. ii the KIIs and FGDs was done in close coordination with the Contracting Officer s Representative (COR), Implementing Partner (IP), and relevant government and CSO partners. MAIN FINDINGS Evaluation Question 1 EQ 1: How effective was the ToC in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved? Summary of Key Findings • Overall, the ToC was effective. According to KIIs with DAI staff and USAID partners, 94 percent (12 male and six female) of those interviewed stated that the ToC was effective in identifying the community elements most impacted by VE, as reflected by the pillars of the GoK’s strategy to counter violent extremism, but there were challenges in establishing clear linkages and coordinated partnerships between national and county efforts. • Both objective areas achieved impact. According to interviews with stakeholders involved in the design and strategy of NIWETU, both objective areas achieved impact, but the lack of coordination, particularly between national and county elements of the project, slowed interventions. A delay in Year 1 in establishing effective relationships with NCTC and the Ministry of the Interior and Coordination (MoI) impacted interventions that required GoK participation at the county and national levels. According to KII participants, NIWETU and USAID had a tenuous relationship with the MOI and NCTC as there was no initial agreement letter between USAID and the Ministry when designing the NIWETU SOW which caused some tension with NCTC. This coordination issue filtered down to county governments and in some cases was also noted by civil society partners. • There were organizational issues that influenced observed outcomes. According to management and field staff interviews as well as CSO grantees managed by NIWETU, certain decision￾making delegation issues impacted field staff authority and implementation. The design of the project was dependent on political will, impacting the ability of the NIWETU field teams to effectively support some interventions. There was no evidence to demonstrate that the organizational structure aligned with the hyper-local context. • Activities envisaged in the ToC contributed to increased understanding of VE. As evidenced in CSO interviews, 82 percent (23 male and 13 female) stated that activities envisaged in the ToC contributed to increased understanding of VE and, in some cases, increased influence on Preventing and Countering Violent Extremism (P/CVE). NIWETU’s subgrantees managed to expand their circles of influence on P/CVE issues, and there was evidence that some efforts have sustained post-NIWETU closure. NIWETU activities increased understanding and awareness of VE both at the community level and within CSOs working on CVE issues. At the county level, Community Engagement Forums (CEFs) were instrumental in sustaining an emphasis on P/CVE, and the Kenya School of Government’s Security Management Institute (KSG/SMI)’s legacy interventions are examples of NIWETU’s impact on the CVE landscape in Kenya. • Field work confirmed that women are essential stakeholders in P/CVE efforts. Based on KIIs, ME&A confirmed that women are essential stakeholders in P/CVE efforts. The ToC emphasized the inclusion of women, and NIWETU did effective integration of women into program elements; but it could have done more if the emphasis had been on identifying specific county influencers and potential change-makers at the hyper-local level early in the project. Women stakeholders we interviewed felt that the role of women is a critical activity element and is less about including women as a quota or as representatives of complex CVE issues that impact women. • NIWETU engaged a diverse set of stakeholders. It adapted participation based on research and learning to incorporate groups that were overlooked. However, a weakness of the project design was focusing on stakeholders that were largely in urban and peri-urban areas. iii Evaluation Question 2 EQ 2: Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to VE under DO 1? How can these interventions be improved? Summary of Key Findings • Subgrants were essential in initiating community-led interventions. Based on KIIs with NIWETU stakeholders, over 86 percent (23 male and 11 female) stated that subgrants were essential in initiating community-led interventions, and these efforts were critical in creating space at the county level for a shared identification of local VE drivers as well as creating space for informal networking. • CSOs capacity-building for P/CVE programs is unique. According to NIWETU, stakeholders’ and CSO grantees’ CSO capacity-building for P/CVE programs is unique and requires greater knowledge of national security architecture and tasks. Local CSOs had a key role in strengthening the relationship between the governments and the community, but the tenuous relationships at the county and national levels regarding perspectives on VE can also be a liability—as partnerships between civil society and security institutions can be strained. • CVE Community Champions were essential for NIWETU. Based on field interviews, over 90 percent (22 male and 10 female) of those interviewed felt the CVE Community Champions were essential for NIWETU. NIWETU’s Champions for Change (C4C) approach to choosing influencers for their P/CVE efforts was successful and strategic. The selection process was participatory and contributed to increased cooperation between community stakeholders. • Strategic communications increased NIWETU’s impact and, in some cases, reach. Based on a review of grant reports, quarterly reports, and KIIs with strategic communications staff and partners, strategic communications increased NIWETU’s impact as well as geographic reach. Strategic communications had impact in countering the recruitment messages by AS. It also expanded the reach and capacity of both the program and its partners (CSOs and C4Cs). • Despite limitations and delays the project contributed to some interventions that laid important groundwork for future CVE efforts. NIWETU, through its research and learning partner Wasafiri, issued seven Task Orders (TOs) to Wasafiri, including: (1) the Complexity Aware Planning (CAP) Cycle Framework; (2) detailed county-level VE actors and opportunities mapping; (3) development of the scope and budget for NIWETU’s Baseline Assessment; (4) the subsequent delivery of the Baseline Assessment; (5) a study on VE recruitment dynamics in Isiolo County; (6) facilitation of a partner learning session; and (7) a rapid assessment of the VE-related dynamics in the Boni Forest Reserve Area. These learning efforts provided input for adaptation and improved intervention design. Evaluation Question 3 EQ 3: To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering VE and advancing its J2SR? What do key counterparts perceive to be the positive and negative aspects of the partnership? Summary of Key Findings • There was mixed evidence on NIWETU’s systems approach. Based on the timing of GoK’s CVE strategy and a range of factors beyond the control of the program, partnerships were sometimes problematic. While some stakeholder partnerships survived the life of the project, others were not able to sustain themselves at the level envisioned in the project documents and in KIIs with NIWETU senior staff. • There was minimal private-sector engagement. Despite evidence of effort to engage, there was minimal private-sector engagement despite this identified need. According to KIIs and project documents, NIWETU attempted to establish private-sector partnerships as envisioned in the project iv design, but the private sector was reticent about being associated with CVE. Based on KIIs, one way of improving private sector engagement would be a component focused on youth mentorship and exposure to business and private sector enterprises. This would allow for a concentrated effort to establish C4C participants that were members of the business community and who also worked as agents of economic improvement within the counties they were connected to. The other suggestion was to have a PPP as a component for each county to engage in with a focus on PVE led by women and youth leadership. A county wide competition was suggested to promote innovate approaches to address public information gaps through a campaign addressing CVE messages. • Some NIWETU inputs continue to be used. NIWETU’s interventions at the county level have been adopted and sustained beyond the life of the project. These include media products, CEFs, and the participation of C4Cs in security coordination meetings. Many of NIWETU’s civil society partnerships were adopted by other donor programmes working on P/CVE. • Weak government coordination had an impact. Based on KIIs, it was evident that weak government coordination had an impact on Objective 1 and Objective 2 outcomes and impacts. There was a lack of clarity about the roles of government stakeholders working on CVE, and this affected capacity to address P/CVE. There was also evidence of tension as to how this was handled between the field and Nairobi. This was particularly acute in the early stages of the project in partnership planning with NCTC. The initial delay in establishing clear lines of coordination with NCTC and the MOI resulted in significant delays. Once NIWETU started working with the KSG SMI intervention some of these coordination issues improved, and by the end of the project they were better. The initial coordination issues continued to plague the project although the relationship between NIWETU, KSG and the Ministry improved through exposure. • CVE is multisectoral, and this was less of an emphasis in planning partnerships. According to stakeholders working in areas impacted by VE, CVE is multisectoral—and this was less of an NIWETU emphasis. NIWETU made limited deliberate efforts to delineate engagement aimed at security actors and civilian government actors. A multisectoral approach was needed to ensure more deliberate interventions for distinct state actors and local interventions. Evaluation Question 4 EQ 4: How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to VE? How can these interventions be improved? Summary of Key Findings • NIWETU identified and included at-risk youth in its interventions. Based on evidence, NIWETU identified and included at-risk youth in its interventions. These included training, media campaigns, and trauma healing. However, the definition of “at-risk” youth varied across activities and counties. According to KIIs, there is a need to change the terminology of “at risk” to address the changing nature of recruitment targets of Al-Shabaab (AS) and address inclusion issues relating to elite youth capture of P/CVE activity interventions and grants. • Rural and remote youth were not as involved. According to stakeholders, rural and remote youth were not as involved. Interventions focused on a few sub-counties, especially in urban and peri￾urban areas, thus neglecting youth in remote areas. Over 80 percent (26 male and 13 female) of CSO and field staff interviews conducted stated this concern as an important consideration for future intervention planning. • Strategic communications interventions increased scope and reach. According to staff and partners, strategic communications interventions increased scope. There was strong evidence that targeted messages were developed, addressing youth at risk; but limited evidence on how effective the medium used was. This was partially due to interruptions in planned activities due to COVID-19 restrictions. However, stakeholders also noted the changing and evolving types of social media platforms used by youth, AS, and communities seeking information on P/CVE issues. v • Most NIWETU interventions focused on prevention. According to DAI staff and stakeholders, 86 percent (12 male and 5 female) stated that most NIWETU interventions focused on prevention and not countering radicalization and recruitment. Youth were included in the development of some CAPs and contributed to the emergence of informal P/CVE networks, although the degree of influence varied by county. Activities targeting reduced radicalization and recruitment largely feel in the prevention space and focused on education. CONCLUSIONS Evaluation Question 1 • NIWETU created entry points for P/CVE partnerships in Kenya. The ToC and design of the project carved out space for CVE to be ‘desensitized’ within community dialogue and a shared perspective on local drivers of VE to be more readily understood by a more varied group of stakeholders. The program’s learning, and adaptation was not as effective in informing response, but the challenges encountered by NIWETU benefited multiple stakeholders and influenced other donor and GoK efforts on P/CVE. There were challenges in establishing clear linkages between national efforts and county level interventions. But the lessons lived/ learned, and the partnerships established have evolved P/CVE programming in Kenya. • Political will is essential in P/CVE design. NIWETU struggled early in the project to establish clear partnerships with NCTC and MoI and this resulted in weak linkages between Objective 1 and 2. Greater awareness by staff of governance issues evident at the county level, along with having clear partnerships with national level entities would have strengthened the ToC. Inconsistent government policy and lack of coordination within the government was a challenge for NIWETU and is a sustained challenge for Kenya’s P/CVE implementation. • Citizen - Government partnerships/ coordination is essential. Partnerships and coordination could have been strengthened by a more systematic/dedicated approach within the program. Over emphasis on Objective 1, was an organizational as well as operational oversight. Despite initial momentum, some CSOs were not able to sustain their interventions due to security, influence, and resource constraints. The ToC did not adequately account for this vulnerability. • NIWETU’s ToC highlighted gaps in P/CVE design. Women are essential stakeholders and partners in P/CVE not only victims or quotas to be included. Their perspectives should be centralized and not strategized. Gender plans and strategies are outdated for P/CVE, particularly in Kenya. Youth at risk is a term that is not relevant in P/CVE. Radicalization and recruitment impact communities and the emphasis on “youth-at-risk” is problematic. There should be a pivot to social inclusion principles and practices to mitigate the impacts of VE on communities. Formal P/CVE networks are essential to safeguard and ensure that civil society has a space where they lead the discussion. These networks can “invite” security elements but should not merge national security issues with human security needs. Evaluation Question 2 • NIWETU discerned the “P” from the “C” in P/CVE. Objective 1 was dominant in the NIWETU landscape. Three of the four result areas were addressed in this space: 1) improved CVE knowledge and skills; 2) community networks established; and 3) community-led strategies and initiatives. Research and learning of NIWETU was broad, ambitious, and useful for future programs, but was too complex in most interventions to be integrated and sustained with local partners and efforts. In P/CVE programs there is a need to have multiple mechanisms for funding initiatives beyond grants. There is a need for more responsive and flexible funding for emerging priorities and needs. Subgrants improved CVE knowledge and skills, contributed to informal networks, and created space for community-led strategies and initiatives. vi • Organization capacity building is unique for P/CVE. NIWETU worked with a multitude of CSOs who have systemic mistrust of security institutions and have designed interventions to mitigate and to collaborate with these issues. CSOs working on P/CVE issues need more education on national security architecture in Kenya regarding VE response. Human security and national security are distinct and NIWETU partners highlighted some of the essential elements necessary to ensure that human security becomes more of a priority for county level partners working on CVE. • NIWETU interventions that increased scope. CVE Champions increased the reach of NIWETU and worked beyond the sub-county level. C4Cs were effective influencers, recognized by the government, community and CSOs. Their ‘hyper-local’ contributions improved the learning and adaptation process of the program within the counties. Community-led interventions such as trauma healing has significant impact on both PVE and CVE. Trauma healing provided a safe space for affected community members to address the impact of VE and their vulnerability at both individual and community level. Strategic communications have the potential to catalyze greater impact beyond the constraints of resource and geographic scope. NIWETU’s activities emphasized the communication and not the strategy. There is a need to adjust these interventions to match the current digital landscape in Kenya. Evaluation Question 3 • NIWETU’s systems approach was a partial success. Systems approach only addresses a partial set of issues relevant to implementing a national P/CVE strategy. To the extent possible, NIWETU contributed to advancing the J2SR principles. Multistakeholder involvement in the development of CAPS and establishment of CEFs established and deepened relationships and contributed to enhanced capacity for preventing and countering VE. NIWETU diversified the landscape of partners working on CVE. Multistakeholder involvement in the development of CAPS and establishment of the KSG/SMI deepened relationships and contributed to enhanced capacity for preventing and countering VE. • P/CVE programs need a multi-sectoral approach to be self-reliant. Private sector partnerships can address socio-economic fault-lines as well as inclusion issues relevant to PVE. There was no clear engagement modality for private sector partnerships. This was a missed opportunity to innovate a role for the PS in P/CVE as well as leverage local resources in support of J2SR. NIWETU identified CVE gap areas that require commitment of GoK across ministries and service sectors. Some NIWETU interventions helped identify gaps in services as well as commitment necessary to adopt P/CVE tools and interventions. (CEFs, Psychosocial needs, Returnees, Amnesty issues etc.). There is a Faultline between human and national security concerns that AS capitalizes. Jurisdictional issues hampered collaboration, commitment, and adoption of NIWETU tools. USAID’s P/CVE programs need to draw clear lines between security interventions and governance interventions to maximize impact and mitigate risk. (CVE toolkit/ NCTC). Evaluation Question 4 • The term “at-risk” can undermine P/CVE programs working on inclusion. NIWETU’s targeting of at-risk youth (male and female) were not always aligned with P/CVE needs at the county and national level. There was a limited focus on youth issues as they relate to CVE contexts (i.e., reintegration, deradicalization, counter-messaging, etc.). The targeting of youth at risk of radicalization and recruitment as an element of a P/CVE program was more effectively done by local CSOs who had a better understanding of the context and risk factors. The issue of returnee reintegration is an area of CVE that does not yet have a ‘home’ in USAID P/CVE efforts. Programs that focus only on ‘at-risk’ sometimes neglect opportunities to assist in other critical effects of VE, such as reintegration of returnees. • Recruitment and radicalization are a community concern. There is increasing evidence that to address radicalization and recruitment of youth, programs should also focus on women-led initiatives seeking to mitigate risks. These efforts tend to be multi-sectoral to address drivers as well vii as risks. Recruitment and radicalization are an urban and rural issue. To address issues of geographic reach, NIWETU worked with the C4Cs, expanding to secondary locations within counties and utilized media partnerships to expand messaging. Although NIWETU’s interventions utilized various communication platforms, in CVE interventions, effectively reaching youth requires an understanding of how youth consume and interact with information. It also requires including communities and families impacted to design relevant content. SUMMARY OF RECOMMENDATIONS Recommendations for P/CVE Design (Relevant to EQ1) Recommendation 1: Design future P/CVE programs with a prescribed governance intervention as well as a clear partner within the National Government. CVE programs require extensive collaboration at the national level. Donors and development actors require clear GoK partnerships to implement P/CVE interventions at the sub-national level. Recommendation 2: Future P/CVE programs should be multisectoral. P/CVE programs should integrate more effectively with other sectoral projects and not only focus on hyper local conditions. VE is a long-term issue. P/CVE activities should be designed to interlink with other development efforts and funding (i.e., education efforts, mental health, and job creation or economic growth/ livelihood programs). Recommendations for Interventions and Learning (Relevant to EQ2) Recommendation 1: Future programs need CVE champions and influencers. Citizen-centric engagements are an effective modality for enhancing community capabilities around P/CVE. Future programming should scale up and dedicate more resources to C4Cs or a similar mechanism. Recommendation 2: Strategic Communication is essential in future P/CVE programs and can also address gender gaps. These interventions need to mature to address the changing social media and digital landscape and to strengthen the voices, perspectives, and leadership of women impacted by VE. Future P/CVE projects should have dedicated funding streams to leverage Strategic Communication efforts and integrate them into field-based activities that aim to enhance reach and demographic scope. Recommendations for Resilience and J2SR (Relevant to EQ3) Recommendation 1: Future programs should address self-reliance early and high as VE is a long-term issue in Kenya. P/CVE interventions and investments should be mainstreamed into the development agenda at county and national level. This would make a systems approach more impactful by emphasizing ownership over partnership. It would also allow economic realities to be aligned with interventions. Recommendation 2: USAID should maintain key leadership engagements with multiple government stakeholders even when there are no active P/CVE programs. The goodwill and momentum developed in NIWETU can be leveraged by a future program to leapfrog existing systemic issues and facilitate a degree of co-creation for future interventions. Recommendations for Future P/CVE Targeting (Relevant to EQ4) Recommendation 1: Future programs should focus on risk factors and not ‘at-risk’ youth. Switch emphasis from efforts targeting ‘at-risk youth’ to designing activities that provide opportunities that enhance inclusion, social cohesion and address dynamic vulnerability factors. Recommendation 2: Returnees, Radicalization and Recruitment are interwoven in Kenya. Future programming should be linked with other initiatives working on DDR. While this is highly dependent on political will, a future program can act as a facilitator for the community component of DDR. Recommendations for Future P/CVE Programs viii Recommendation 1: A future P/CVE program should incorporate a liaison officer seconded to the county governments as well as with NCTC. The liaison officer at the counties would be responsible for improving the relationship and political buy-in from the county leadership as well as facilitating relationships for field and program staff. The liaison officer would have deep insight into both the program’s priorities as well as the county administration’s priorities and contribute this knowledge to the design and implementation of realistic, impactful interventions. These positions can be filled through a joint recruitment by the program and the government. Particular challenges with this approach may include the possible pay disparities between the liaison officer and counterpart civil servants as well as ensuring adherence to USAID and GoK HR guidelines. Recommendation 2: Implement multisectoral programming by co-implementing with other USAID programs, staff, and initiatives. Improving linkages at the field level with other USAID programs and their staff is also crucial. This not only contributes to improved analysis and understanding of the political economy but also provides to cross-leverage opportunities for learning, co-design, and co￾investments. Improving CVE Programming in Development The current USAID Policy for Countering Violent Extremism in Development Assistance2 does not include a clear delineation between programs that address VE drivers, those that reduce risk, and those that seek to build CVE capability in the countries in which USAID works. A clear understanding and description of those differences is important for mission to understand how both violent extremism (the problem) and countering violent extremism (the means to solve the problem) might be assessed and addressed in development planning and intervention design. There are essentially two categories of CVE programming: (1) VE Risk Reduction Programming and (2) CVE Systems Programming. Risk Reduction Programming address the reasons that people find their way to radicalization and violent extremism. These programs address the drivers, particularly the push and pull factors3, which are context specific and vary by country and location. These are the factors that reduce the likelihood or risk that individuals might convert to VE or engage with VEOs to recruit or radicalize vulnerable populations. As we have seen in the evolution of AS’s tactics to radicalize and recruit, they no longer merely seek youth, but also teachers, coaches, and influential members of communities in which they operate. They often seek secondary influencers to recruit youth and other members into their movement. In application, Risk Reduction Programs: • Currently account for most of the programming targeting VE within USAID (central and mission programs). Most of these programs are PVE and not CVE programs, despite the classification in the contract or in USAID branding. • Address the context specific drivers of VE and have end goals that reduce the presence of these drivers in targeted geographical areas. These programs will always require hyper local approaches with enhanced assessment and PEA. • These programs emphasize targeting people identified to be at-risk of radicalization and recruitment. These targets are critical in achieving success and often are under-targeted by not emphasizing the role of women and community influencers in recruitment and radicalization strategies of VEOs. • USAID and development programs in this category at best are ‘CVE sensitive’ and are primarily focused on prevention or PVE. There is often weak reporting and measurement methods for tracking reduction in radicalization or decreases in recruitment. 2 USAID Policy for Countering Violent Extremism in Development Assistance, USAID April 2020 3 The Development Response to Violent Extremism and Insurgency, USAID 2011. ix • Most programs also focus on reducing risk but fail to identify ways to articulate changes in the context that can be directly attributed to development program interventions. There is a need in these programs to have an independent body work alongside the IP to conduct baselines, midlines, and end lines to assess, track and document contextual and behavior change to establish risk reduction in factors that lead to engagement with VEOs. CVE Systems Programming build a nations and community’s ability to recognize and intervene to prevent and counter VE. These programs do not address drivers; instead, these programs work to create CVE policies, strategies, and institutions in partner governments. They seek to train and prepare government officials to recognize and understand and work with citizens and local communities to prevent VE. These programs work at the national, sub-national, and local level. Populations targeted are inclusive and include CSOs, teachers, religious leaders, special interest groups, the private sector, parent-teacher associations, community leaders, and advocacy groups to not only recognize the signs of VE but also to have the agency, authority, and power to intervene in the process. In application, CVE Systems Programming: • Accounts for a very small fraction of current USAID programming. NIWETU was largely a PVE program with some elements of CVE Systems Programming. • These programs do not address the drivers of VE; instead, programming seek to empower and activate national, sub￾national and local governments to empower communities to engage and lead CVE efforts. • These programs work on strengthening institutional capacity and community capabilities to address VE by creating systems to cooperate, partner, and collaborate on critical intervention needs. NIWETU did not require a ‘systems approach’ but rather, a ‘systems way of thinking’ about CVE needs and targets. • CVE Systems Programs do not seek to target ‘at-risk’ populations but instead improve safety nets, social protection systems, and address systemic grievances and critical needs to strengthen community and sub-national response to VE and VEOs. These programs may seek to increase access to livelihoods and critical services to reduce the vulnerability of communities to engage and support VEOs seeking to operate in their locales. • These programs intentionally seek to build frameworks, structures, and systems that can operate with local resources (social capital and funding) and that directly engage current budget and organization constraints of national and sub-national partner institutions. KSG/SMI was a ‘legacy intervention” of NIWETU and reflects a CVE Systems approach to CVE programming. This is one of the few interventions within NIWETU that fall into this category. • CVE Systems Programming requires national collaboration through established MOUs, or partnership with critical national or sub-national structures. The timing of NIWETU was appropriate as Kenya had a new National CVE Policy, but some of the interventions failed due to weak ownership and co￾creation with NCTC and MoI. When the project attempted to have partners adopt a CVE toolkit it was rejected by NCTC as it was felt the process did not reflect the needs of NCTC and the national CVE landscape. As the toolkit process was largely done with local partners (CSOs and stakeholders) it was oriented towards civil society and not towards shared spaces of partnership and implementation that were necessary at the sub-national level to implement the national CVE policy. • CVE Systems Programming requires working with elements of civil society that engage or need to engage on national security and human security issues. There is a grey space that exists in CVE that requires development partnership as the interventions end goal is to improve community security and “The CVE toolkit is an intervention that did not work. As much as VE is VE wherever you go, there was lack of contextualization during the development. There was a feeling among different stakeholders that the toolkit did not speak to the Kenyan context. In future, an external organization like HEDAYA who were leading in the development of the toolkit could review the document but not design it. We saw interventions that were led by local organizations having better understanding of the local issues. A good example is what happened in Kamukunji where KACPEN researched on the VE context, understand the context, and came up with an action plan relevant to the community.” ~ Female KII, DAI Management x well-being. Countering VE requires community support and sustained trust between citizens and national security institutions. There is a critical need to educate civil society and constituent populations about the role of security institutions. There is a critical need to strengthen the role of citizen engagement in oversight of sub-national security institutions to ensure that this becomes an enduring role. 1 1.0 EVALUATION PURPOSE AND QUESTIONS 1.1 EVALUATION PURPOSE The purpose of this evaluation is to measure the effectiveness of the United States Agency for International Development/Kenya and East Africa (USAID/KEA) s Kenya NiWajibu Wetu (NIWETU) Activity in achieving its key objectives: (1) Community mobilization to address Violent Extremism (VE) enhanced and (2) Government responsiveness to VE improved. NIWETU, a four-year (2016–2020) activity, was designed to improve the capabilities of communities, subgrantees, and the Government of Kenya (GoK) to identify and respond to VE threats at the national and county levels with a focus on five core counties.4 The Mission initiated this final performance evaluation to provide a systematic and objective assessment of NIWETU s performance, outcomes, and lessons learned. The results of this evaluation will be used to highlight lessons and promote learning of effective development approaches that merit replicating in Countering Violent Extremism (CVE), a relatively new program area across the Agency and in the peace and security sector. With NIWETU’s mid- and end-line evaluations and its final report, the results of this evaluation will be used to inform the design of the Mission’s forthcoming CVE activity. 1.2 EVALUATION AUDIENCE The main audience for this evaluation will be USAID/KEA staff, who will use the findings to inform ongoing interventions designed to mitigate conflict and the design of new programs. Evaluation results will also be shared with a secondary audience including USAID/Washington, USAID/East Africa, and other key stakeholders, including GoK ministries, departments, and agencies (in particular, the National Counter Terrorism Center (NCTC), Ministry of Interior (MoI), Coordination of National Government, and select county governments) and development partners. The evaluation was designed to provide an informed and evidence-based high-level overview of the program’s progress and achievements with pragmatic recommendations for the way forward. 1.3 EVALUATION QUESTIONS USAID/KEA designed four main Evaluation Questions (EQs) to assess whether NIWETU had achieved its main objectives. The questions were: . The questions were: 1. How effective was the Theory of Change (ToC) in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved?’ 2. Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to violent extremism under Objective 1? How can these interventions be improved? 3. To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering violent extremism and advance its J2SR? What do key counterparts perceive to be the positive and negative aspects of the partnership? 4. How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to violent extremism? How can these interventions be improved? 4 Garissa, Isiolo, Mandera, Nairobi, and Wajir. 2 2.0 BACKGROUND NIWETU s technical approach was grounded in Kenya s National Strategy to Counter Violent Extremism that promotes participation of an inclusive” array of government and non-government stakeholders in CVE efforts and evolved to pursue a novel systems-based” methodology that informed USAID s 2020 CVE policy. NIWETU supported the USAID/KEA Country Development Cooperation Strategy (CDCS) 2016–2020, Development Objective 1 (DO 1): Devolution effectively implemented,” and its Intermediate Result (IR) and sub-IRs.5 The goal of NIWETU was to Improve CVE capabilities to identify and respond to VE threats” targeting the five Kenya counties most vulnerable to VE that included Garissa, Isiolo, Mandera, Nairobi, and Wajir. To realize this, NIWETU s conceptual framework was organized under two objectives: 1) Community mobilization to address VE enhanced;” and 2) Government responsiveness to VE improved.” Figure 1: NIWETU Goals, Objectives, and Key Activities The NIWETU ToC stresses the interconnectedness of the four result areas that converge to accomplish the program s objectives: 1) Improved CVE knowledge and skills; 2) Community networks established; 3) Community-led strategies and initiatives; and 4) Increased government capacity. The following ToC is designed to apply the NIWETU systems-based approach: IF a wide variety of local stakeholders—women and youth included—have an enhanced understanding of VE and their leadership skills are strengthened, THEN stakeholders will expand their circle of influence and be able to generate community engagement and support for CVE efforts. And, IF CVE stakeholders, organizations, and leaders work together to analyze local VE dynamics and prioritize CVE activities in their contexts and intentionally establish/strengthen functional coordination, THEN community CVE networks can emerge. And, IF CVE networks are established and harness unique perspectives and skills of a diverse cross-section of community stakeholders, THEN the community will mobilize to develop and implement initiatives that address local drivers of VE, thereby effectively reducing VE. And, IF targeted national and local government officials’ CVE capacity is enhanced, THEN national and county governments will better understand multiple causal layers of VE and effectively collaborate with communities to develop and implement responsive CVE policies and strategies. 5 IR 1.3: “Citizens participate in county affairs;” and its sub-IRs: 1.3.1: “Capacity for civic engagement improved;” 1.3.2: “Women, youth, and marginalized groups participation and representation enhanced;” and 1.3.4: “Conflict mitigation mechanisms at the county level strengthened.” 3 NIWETU was designed to tackle radicalization processes by strengthening an intimate understanding of specific vulnerabilities within communities and institutions at the county and national levels. NIWETU took a community-led approach to CVE, focusing on community dynamics in each VE hotspot. NIWETU strengthens each community s ability to withstand the influence of Violent Extremist Organizations (VEOs) and reduce overall VE activity. NIWETU largely worked in the education and prevention space. (See Figure 2.) While these stages depict a linear progression an individual, family, or community may take into radicalization, there are also cases of forced recruitment occurring in Kenya. NIWETU worked with communities to analyze and then apply knowledge to the specific dynamics in each case. Figure 2: Updated NIWETU Conceptual Framework 2019 Ultimate achievement of NIWETU s objectives rested on several key assumptions: ● Government and community willingness to engage with NIWETU activities, and a willingness to work together to address VE in hotspot counties and at the national level, particularly with key national-level partners (i.e., MoI, NCTC, and KSG). ● Level of VE did not surpass the government or community’s ability to maintain an operational environment. ● USAID support for a flexible, adaptive approach to programming to respond to changing VE dynamics, address rapid-response needs, and revise Monitoring and Evaluation (M&E) plans and targets based on learning and the effects of programming. NIWETU began community and government work in Garissa, Isiolo, Nairobi, and Wajir counties from the start of Year 1, with further engagement at the national and inter-governmental levels. In Year 2, the geographic scope expanded to reflect Scenario B s addition of Mandera County. In Year 3, NIWETU expanded to work across 37 counties, in partnership with NCTC, completing Rapid County Action Plans (RCAPs) in each county in Kenya. With approval from USAID, NIWETU had authorization to work in all counties in Kenya. This evaluation targeted the initial counties from Year 1 and Mandera locations with some of the earliest interventions. 4 3.0 EVALUATION METHODS AND LIMITATIONS 3.1 EVALUATION METHODOLOGY To research evidence of progress and achievements in CVE, the ET conducted a participatory and mixed￾methods design. The ToC, along with its contextual assumptions, provided the framework for the evaluation design. The ET reviewed secondary sources, gathered largely qualitative primary data through semi-structured KIIs and FGDs and, also, extensively used—as appropriate and relevant—data from NIWETU’s Monitoring, Evaluation, and Learning (MEL) system (monitoring data, reports, baseline and endline surveys) and secondary data reviews. 3.2 DATA COLLECTION The ET employed a purposive sampling approach to ensure the inclusion of diverse beneficiary and stakeholder groups. The evaluation methodology ensured stakeholder mapping and sampling that was representative of the approach and partners of NIWETU in the five target counties,6 with key stakeholder groups including national and county-level government counterparts, civil society, and local l Figure 3: Stakeholder Type and Proportion partner organizations (CSO grantees and partners such as KSG and NCTC), USAID/KEA, and DAI leadership. The ET conducted 80 KIIs both virtually and in-person (as well as five FGDs that were comprised of beneficiary populations of NIWETU CSO grants). The ET did interviews in Mandera entirely remotely due to security concerns. They ensured that marginalized groups such as women, youth, minorities, and other vulnerable populations were represented. Of the KIIs they conducted, 26 percent were women, 74 percent were men, and 38 percent were youth. The main goal of purposive sampling was to recruit the most knowledgeable partners and stakeholders to best answer each of the EQs. The ET selected participants for the KIIs and FGDs in close coordination with the COR, IP, and relevant government and CSO partners. See Annexes 3 for the ET s Interview List. 6 Considering the similarity of communities and the security situation in Garissa, Mandera, and Wajir counties, the ET plans to sample Garissa and Wajir counties, and, if needed, will cover Mandera through remote interviews with key beneficiaries and stakeholders. 5 3.3.1 Data Collection The ET used qualitative and quantitative data collection methods from primary and secondary sources tailored to answer each EQ. They recognized that relying solely on one method or source of data has its own limitations and, as such, they triangulated using different sources and methods. To the extent applicable, beyond the project MEL data, the ET reviewed available studies and surveys by government or other donors. In the following sections, we discuss the data collection methods and tools that are aligned with the EQs and balance methodological rigor and feasibility. 3.3.2 Data Tools Document review: This involved reviewing key documents identified in the Statement of Work (SOW) as well as independent literature review of published and unpublished articles, surveys, and CVE-related resources and publications from government and Non-Governmental Organizations (NGOs) in Kenya and East Africa. To this end, to guide the desk review and extraction of quantitative and qualitative secondary data, the ET created a comprehensive list of available documents relevant to answering each of the EQs. Illustrative NIWETU-related documents7 and reports included quarterly and annual reports, baseline and endline survey data and reports, and external research and studies produced by government and research organizations as well as donors supporting CVE and conflict mitigation programs. The desk review identified what data and information were already available, and gaps that needed to be filled through primary and further secondary data reviews to ensure that each EQ was answered completely. Annex 4 has a complete list of documents identified for review. Primary data collection: This was conducted with a range of project beneficiaries and stakeholders at the national, county, and community levels using tailored data collection tools. (See Annex 2.) Information gathered deepened the ET’s understanding of key internal and external factors affecting interventions, implementation, and results—as well as gauging how stakeholders perceived the performance of the project and its efficiency, effectiveness, and sustainability. Primary qualitative data offered the opportunity to understand the quality of project interventions, perceptions of CVE capacity-building, partnership, coordination and collaboration, challenges, gaps, and lessons learned. The ET undertook data collection from government organizations, local authorities, community leaders, knowledgeable community members, beneficiaries, and youth from the focus counties (see Tables 1 and 2). As security and COVID-19 restrictions limited access, the ET sought to do as many KIIs as possible face to face; however, some were done remotely using Zoom, Skype, phone, or WhatsApp. To ensure inclusivity, they often conducted phone interviews more than once to accommodate follow-up concerns or questions, if necessary. Inclusion: The ET focused on tools to increase inclusion. They mainstreamed gender in all analyses and applied tools to not only interview a quota of women but, also, to address and follow up on issues specific to women and CVE. They included interview prompts to ensure that they targeted female interviewees and concerns to ensure a “more than” balanced perspective. They also included youth perspectives and employed interview prompts to ensure greater understanding of youth concerns and issues. They used triangulated analysis to ensure that responses were not only “gendered” but that they analyzed the application of gender tools, approaches, and the implementation and outcomes of strategies. a. KIIs: The ET conducted KIIs with the key stakeholder groups listed in the SOW and additional stakeholders identified through the interview process itself. KII participants included 7 Additional documents include USAID/KEA Country Development Cooperation Strategy (2016–2020) and Performance Management Plan (including Performance Indicator Reference Sheets); Kenya’s National Strategy to Counter Violent Extremism, September 2016; USAID’s Policy for Countering Violent Extremism Through Development, April 2020; and County Action Plans for NIWETU’s five core counties. 6 USAID/KEA staff, NIWETU staff, national and county government officials, other donors, CSO subgrantees, community leaders, religious leaders, women, and youth. The ET prepared and employed semi-structured discussion guides aimed at eliciting responses and feedback on CVE to answer the EQs. Table 1: KII Participants by Stakeholder Group, County, and Gender Location USAID/ KEA Staff DAI Leadership Staff National Government County-Level Government Civil Society Total Male Female Virtual 6 10 3 X X 19 14 5 Nairobi 4 2 5 3 5 19 16 3 Isiolo X 2 X 3 5 10 6 4 Garissa X 2 X 3 9 14 12 2 Wajir X 2 X 3 5 10 7 3 Mandera (virtual) X 2 X 2 4 8 7 1 TOTAL 10 20 10 15 25 80 62 18 FGDs: FGDs specifically targeted beneficiaries, CSO subgrantees, women, youth, and community members on their experiences and opinions. The ET selected FGD participants in close consultation with the IP and government and CSO partners, as well as CSO subgrantees. They completed five FGDs with beneficiaries and CSO subgrantees using a set of discussion guides adapted to each beneficiary group. To ensure meaningful discussion, FGD participants were as homogenous as possible (e.g., organized by gender or activity role), where appropriate and feasible, and the FGDs took place in venues where participants felt safe and secure. As the project had ended in 2020, FGDs were challenging and were primarily arranged in coordination with CSO grantees. FGDs were primarily comprised of beneficiary groups of CSO grantees. They included an all-male FGD and all-female FGD in in Kamukunji, a FGD of KSG/Security Management Institute Training of Trainers (ToT) staff, an Isiola FGD, one comprised of all youth, and another of CEF members from Wajir. The ET built the FGDs in close collaboration with CSO grantees at the county level to ensure knowledge and exposure to NIWETU interventions. Table 2: FGD Participants by County Location FGDs Total Male Female Nairobi in (Kamukunji) 2 10 6 4 Isiolo 1 3 1 2 Kenya School of Government (KSG) Training of Trainers (ToT) 1 3 2 1 Wajir 1 3 2 1 TOTAL 5 19 11 8 7 3.3.3 Data Analysis and Reporting The ET applied analytical approaches specific to the different data sources and methods, followed by triangulation and interpretation to generate evidence-based findings, conclusions, and recommendations. Analysis of data from KIIs and FGDs was primarily content analysis, using qualitative data analysis software and thematic analysis. Content analysis is a well-established, credible, and systematic technique that compresses words from the text into fewer content categories based on explicit rules of coding data from qualitative interviews, secondary sources, and relevant documents. Based on the initial review of raw data and applying a combination of deductive and inductive coding, the ET developed an initial codebook and added emergent codes to the codebook as data collection and preliminary analysis progressed. After coding the interview transcripts and documents, they began the analytic process by identifying patterns in the data, categorizing them into themes and sub-themes, grouping and regrouping, and linking them with the EQs. This allowed them to synthesize triangulated findings and draw inferences from the data by objective and systematic analysis of the data by EQ. 3.3.4 Evaluation Limitations and Mitigations NIWETU closed two years before this evaluation, which presented some unique challenges. This evaluation relied primarily on qualitative primary data collection. Limitations included biases due to purposive selection of key informants for KIIs and FGDs, which does not necessarily reflect the views of the majority; self-selection bias; affiliation with NIWETU; socio-demographic characteristics, experience, and unverifiable perceptions; and opinions about changes related to CVE effectiveness and engagement and government responsiveness. Also, while the ET believes its purposive sampling approach with predefined selection criteria was sufficiently rigorous for the purposes of this evaluation, it is possible that it was not able to adequately represent the full range of stakeholder opinions. The other contextual challenges included security in some of the counties (Garissa and Mandera) and the ongoing COVID-19 pandemic, which posed travel restrictions that the team had to adapt to. To minimize biases, limitations, and challenges, the ET applied a range of strategies, including investing time explaining the purpose of the evaluation. The Team Leader (TL) organized periodic check-ins to verify data completeness, consistency, and accuracy—and ensured that the team identified and timely corrected any anomalies. The regular communication with guidance and feedback from the COR was critical to address emerging challenges. The ET also implemented the following mitigation strategies: • Coordinated closely with the COR and IP to facilitate introductions and mobilization of stakeholders to ensure wider participation and keep participants well informed of the evaluation’s purpose. Because the project had ended two years before, there were challenges in obtaining KIIs with knowledgeable stakeholders. • Implemented a range of measures to minimize the impact of COVID-19 by adhering to social distancing, wearing masks and gloves, following proper hygiene practices, and avoiding large gatherings. The team also did some KIIs remotely to accommodate stakeholder perceptions of risk. • Took the necessary precautionary security measures, including keeping a low profile; following USAID and government travel and security advisories; and, through the local ET members who know the study areas well, gathering information on the security situation from established contacts before and during data collection to avoid security risks. • To the extent possible, carefully reviewed secondary data and documented any known data limitations of the NIWETU activity. • Performed rigorous data triangulation from multiple sources and methods and conducted numerous team analytical exercises to ensure data and evidence integrity. 8 4.0 FINDINGS, CONCLUSIONS, AND RECOMMENDATIONS 4.1 FINDINGS 4.1.1 Evaluation Question 1 EQ 1: How effective was the ToC in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved? Definitions: “Effective,” defined as sufficiently designed to ensure that aspects of all four objective areas of NIWETU were addressed. Evidence The TOC was effective in increasing awareness of VE and was specifically credited for helping to ‘de￾stigmatize’ community conversations about VE. This, according to participants, emboldened community leaders to speak publicly against recruitment and promote awareness on identifying signs of radicalization. Community leaders advocated for and utilized airtime on both community and private radio stations. In Wajir, a female leader noted that on three separate occasions, they had managed to intercept the recruitment of youth including one incident that involved security officers pursuing a vehicle heading to the Somali border. Overall, participants consistently noted that a lasting impact of the program was the consistent community dialogue that emerged as a result of the design and activities under the TOC. According to KIIs conducted in the field, although the project ToC was effective in addressing the need to increase awareness of VE, the majority (67 percent; 17 male and five female) of CSO grantee partners interviewed also noted that NIWETU was not flexible enough to support the continuity of established interventions and partnerships. Complex and sustained tensions between community groups and security personnel and systems often made interventions powerful but temporary. According to senior DAI staff and USAID counterparts interviewed, the ToC was effective in identifying the community elements most impacted by the radicalization process, but some elements—including those under Objective 1 (community capacity)—were impacted by the time allocated to specific partnerships. A delay in Year 1 in establishing effective relationships with NCTC and MoI impacted aspects of the original intent of NIWETU. The delay in partnering with NCTC created some tensions with key personnel within MoI and impacted cooperation with NCTC at the county level. Based on a review of the MEL documents as well as KIIs with GoK officials, there was evidence that NIWETU’s interventions under Objective 2 contributed to improved government understanding and capacity to deal with VE. The design of the project was dependent on political will, and this impacted the ability of the field team to coordinate with county officials and in some circumstances plan training and field events where GoK participation was critical. The partnership with KSG/SMI was critical to address this, as it helped to create stronger relationships between DAI staff and county and ministry officials. There was limited evidence to demonstrate that the organizational structure aligned with the hyper-local context. Although field staff were hired from the targeted counties, there was an expressed need in interviews with staff for greater access to strategic planning in the field and skills to engage with political will issues. Based on interviews with DAI and USAID, there was evidence that the ToC was not uniformly understood by the field staff and the management team and, in some circumstances, there were varied “In the development of the Sub-County plan of action, we mobilized families from the 14 villages in Majengo. Facilitators were attached to the groups where there was a chairperson and a secretary. They identified the issues affecting their villages and aligned them with the national strategy on VE. Some of the issues were on women, youth and even ideology.” – Male KII, CSO Grantee, Isiolo 9 interpretations of how to target key populations. Interviews indicated that this impacted both the operations as well as local partnerships. Based on field interviews and a review of quarterly and annual reports, activities envisaged in the ToC contributed to increased understanding and in some cases, improved leadership on P/CVE—both with youth and women. CSOs interviewed stated that they expanded their circles of influence on P/CVE issues but also wanted more agency for women—who needed more leadership experience to influence the course of P/CVE planning at the county level. A review of project documentation as well as interviews with CSOs working on gender showed that the ToC’s design, which emphasized the inclusion of women, missed critical opportunities to emphasize the critical role of women in ensuring sustainability beyond the life of the project (e.g., psychosocial support in Nairobi and Isiolo, early warning signs of radicalization). Respondents overwhelmingly agreed that women’s organizations should collaborate more with other organizations working on CVE programming within the four counties of research. In Nairobi and Isiolo, 90 percent of women interviewed (19 out of 21) stated as much, while in Garissa and Wajir, 80 percent (eight out of 10) and 92 percent (11 out of 12), respectively, stated the same. This demonstrates that while women’s involvement in such activities within the community is currently limited, stakeholders do see value in increasing their involvement in these conversations. Nevertheless, the barriers that prevent the wider engagement of women at this level, as discussed in this section, present challenges to successfully increasing the involvement of women’s organizations in CVE discussions and programming. The space for women to contribute to CVE work at the level of community leadership and civil society contains important entry points but remains constrained by gender norms in many communities. While Community-Based Organizations (CBOs) and CSOs comprised largely or entirely of women can be found across all four counties, there is a danger that their work becomes siloed as simply “women’s issues” and is unable to penetrate the wider public space or influence public dialogue on security. Most stakeholders (73 percent; 44 male and 29 female) interviewed stated that NIWETU’s efforts in increasing understanding and awareness of VE was an attribute of NIWETU. CEFs were instrumental in sustaining an emphasis on P/CVE. Evidence showed that NIWETU engaged a diverse set of stakeholders including women and youth and that participation was adapted based on research and learning to incorporate groups that were overlooked, particularly in interventions such as the CEFs and CAPs. However, one of the weaknesses of the project design was focusing on stakeholders that were largely in urban and peri-urban areas. Based on interviews with CSOs as well as triangulating NIWETU interventions documented in annual reports, there is evidence that in some areas, community networks initiated effective P/CVE strategies that went beyond the life of the project. These included information-sharing platforms such as WhatsApp groups in Wajir, CEFs at the county level, and informal networks established for trauma healing. Evidence from interviews showed that networks were established in most counties that have survived beyond the life of the project. There was limited evidence of the establishment and impact of formal networks, but there was evidence that NIWETU’s activities contributed to the emergence of informal networks that continue to engage on CVE. One of the primary issues that emerged in investigating EQ1 was the distinct role of prevention and CVE and how this difference impacted targeted populations such as women and youth. There was an expressed need to look at the value of women and youth leadership and roles in prevention activities—and the need for institutional partnership and coordination in addressing CVE, particularly to reduce radicalization. Key Findings • Overall, the ToC was effective. According to interviews with DAI staff and USAID partners, 94 percent (12 male and 6 female) of those interviewed stated that the ToC was effective in identifying 10 the community elements most impacted by VE, as reflected by the pillars of GoK’s strategy to CVE, but there were challenges in establishing clear linkages and coordinated partnerships between national and county efforts. • Both objective areas achieved impact. According to interviews with stakeholders involved in the design and strategy of NIWETU, both objective areas achieved impact, but the lack of coordination, particularly between national and county elements of the project, slowed interventions. A delay in Year 1 in establishing effective relationships with NCTC and MoI impacted interventions that required GoK participation at the county and national levels. • There were organizational issues that influenced observed outcomes. According to management and field staff interviews as well as CSO grantees managed by NIWETU, there were issues of decision-making delegation that impacted field staff authority and implementation. The design of the project was dependent on political will, impacting the ability of the NIWETU field teams to effectively support some interventions. There was no evidence to demonstrate that the organizational structure aligned with the hyper-local context. • Activities envisaged in the ToC contributed to increased understanding of VE. As evidenced in CSO interviews, 82 percent (23 male and 13 female) stated that activities envisaged in the ToC contributed to increased understanding of VE and, in some cases, increased influence on P/CVE. NIWETU’s subgrantees managed to expand their circles of influence on P/CVE issues, and there was evidence that some efforts have been sustained post-NIWETU closure. NIWETU activities increased understanding and awareness of VE both at the community level and within CSOs working on CVE issues. At the county level, CEFs were instrumental in sustaining an emphasis on P/CVE and KSG/SMI are legacy interventions that are examples of NIWETU’s impact on the CVE landscape in Kenya. • Some interventions were essential in achieving sustained impact. According to interviews with NIWETU stakeholders, interventions such as local and regional networking efforts are essential in achieving sustained impact. C4Cs were effective leaders in strengthening informal networks at a hyper-local level, but the interventions missed an opportunity to support the formation of more formalized networks that could operate at scale. There were also gaps in the inclusion of women, because there was less of a concerted effort to identify local women leaders and change-makers at the county level. • Field work confirmed that women are essential stakeholders in P/CVE efforts. Based on interviews, it was confirmed that women are essential stakeholders in P/CVE efforts. The ToC emphasized the inclusion of women; and NIWETU did effective integration of women into program elements but could have done more if the emphasis had been on identifying specific county influencers and potential change-makers at the hyper-local level early in the project. Women stakeholders that the ET interviewed felt the role of women is a critical activity element and is less about including women as a quota or as representatives of complex CVE issues that impact women. • NIWETU engaged a diverse set of stakeholders. Participation was adapted based on research and learning to incorporate groups that were overlooked. However, a weakness of the project design was focusing on stakeholders that were largely in urban and peri-urban areas. • There was mixed evidence on how well coordination happened between ToC elements. Tensions between security institutions and communities persisted and undermined the will to coordinate between CSO interventions and security partners. County government officials were also reticent to participate in some planned activities and training, although they were more inclined to participate in activities that provided political platforms. 4.1.2 Evaluation Question 2 EQ 2: Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to VE under DO 1? How can these interventions be improved? 11 Definitions: “Effective,” defined as sufficiently designed to ensure that DO 1 was addressed in NIWETU interventions. Evidence Based on project documentation (MEL data, annual reports and Research and Learning Reports) as well as interviews, NIWETU had specific activities designed to enhance community capabilities to identify and respond to VE. These activities were designed also to allow for the engagement of diverse stakeholders necessary to understand and respond to hyper-local conditions impacted by VE. These activities had various levels of impact, and some stakeholders interviewed also noted the distinction of these activities in contributing to prevention as opposed to countering ongoing concerns and VE activity at the local level. There were essentially six activity work streams: 1) CAP, 2) Subgrants to local CSOs and partners, 3) Capacity Building, 4) CVE Community Champions – Network Development, 5) Strategic Communications, and 6) Rapid Response Roster. The CAP Cycle, designed to allow for integrated and adaptive learning and research, was effective in some cases at the USAID and State Department level, but its impacts on field-level planning and response was less evident to stakeholders interviewed in the targeted counties. The field staff interviewed stated that the CAP cycle was more of a reporting tool than one used at the field level. Interviews with field staff of the IP illustrated a tension between this analytical tool to improve local-level response and field-level needs to ensure that communities had more effective partnerships with government and security institutions to ensure a shared understanding of VE risks. Over 85 percent (six male and four female) of field staff interviewed wanted greater responsibility at the field level for political analysis and more staff working on community access to information on CVE and PVE from government partners at the county level. Evidence from interviews showed that subgrants were effective in enhancing community capabilities to identify and respond to VE. This was evident in over 95 percent (24 male and 11 female) of the CSO interviews conducted. Evidence from the KIIs and FGDs with CSO grantees demonstrated that CSOs had a key role in strengthening the relationship between the governments and the community as part of their interventions under the subgrants. However, the short nature of the grants and the systemic challenges in trust between CSOs and government partners was sometimes a challenge in making interventions more embedded or sustained. This trust was particularly problematic in seeking to strengthen partnerships with local security partners. Evidence from both interviews and document reviews shows that NIWETU was successful and strategic at selecting the right persons to serve as C4C Champions. The selection process was participatory and contributed to increased cooperation between community stakeholders. C4C members have continued their work even after the end of the project—acting as a community-government liaison and providing psychosocial support and advocacy support for CSOs. C4Cs continue to support the establishment and activities of informal networks, e.g., youth football clubs, women’s table banking groups, PVE education efforts, and youth engagement activities. Evidence demonstrates that trauma healing was recognized by communities and other stakeholders as necessary for strengthening their resilience against VE and enhancing community cohesion. There was consensus from the interviews that using locally available facilitators/counselors to deliver trauma healing sessions made it more impactful. However, there was a gap in referral mechanisms for more serious mental health issues. One of the areas that was recurring in interviews was the fact that police and other security agencies were dealing with significant trauma from their work as frontline actors, and this trauma impacted community and household relations. “NIWETU gave the facilitators/ volunteers improved ways to express themselves as C4C’s, which then led to people in the community having the courage to speak out on CVE issues. Before, there was a lot of fear around this issue. They feared being targeted by AS, and by the police. They were also traumatized and were not in a good place to network with others until they felt there were people who understood their dilemma.” - Female KII, Garissa 12 There was evidence that strategic communications had impact in countering the AS recruitment messages. Evidence from KIIs indicates that it expanded the reach and capacity of both the program and its partners (CSOs and C4Cs). There was limited evidence demonstrating the effectiveness and impact of the rapid￾response roster and how it helped staff respond to and identify gaps in the field. Rapid response was addressed through a roster of 30 technical consultants capable of working in many CVE spaces and with specific geographic expertise. This roster was employed to conduct research, but staff interviewed did not always feel it was as effective as it could have been. There were also many learning efforts that assisted NIWETU and USAID in identifying program planning and strategy to improve community awareness of VE, but the communication between the field staff and senior staff lagged—causing tensions over priorities in planning for implementation that aligned with grant partners and their abilities to execute identified needs. Key Findings • The CAP Cycle was too complex for the field. According to NIWETU staff, 76 percent (16 male and seven female) of those interviewed stated that the CAP Cycle was overly complex and did not always allow for flexible response. Adaptation and learning from field partners and challenges encountered in political will were overshadowed at times by overly complicated analysis. P/CVE partners under Objective 1 wanted more coordinated sharing across counties and VE experience. • Subgrants were essential in initiating community-led interventions. Based on interviews with NIWETU stakeholders, over 86 percent (23 male and 11 female) stated that subgrants were essential in initiating community-led interventions, and these efforts were critical in creating space 1) at the county level for a shared identification of local VE drivers, and 2) for informal networking. • CSOs’ capacity-building for P/CVE programs is unique. According to NIWETU stakeholders and CSO grantees, CSO capacity-building for P/CVE programs is unique and requires greater knowledge of national security architecture and tasks. Local CSOs had a key role in strengthening the relationship between the governments and the community, but the tenuous relationships at the county and national levels regarding perspectives on VE can also be a liability as partnerships between civil society and security institutions can be strained. • CVE Community Champions were essential for NIWETU. Based on field interviews, over 90 percent (22 male and 10 female) of those interviewed felt the CVE Community Champions were essential for NIWETU. NIWETU’s C4C approach to choosing influencers for their P/CVE efforts was successful and strategic. The selection process was participatory and contributed to increased cooperation between community stakeholders. • Strategic communications increased NIWETU’s impact and, in some cases, reach. Based on a review of grant reports, quarterly reports, and interviews with strategic communications staff and partners, strategic communications increased NIWETU’s impact as well as geographic reach. Strategic communications had impact in countering the AS recruitment messages. It also expanded the reach and capacity of both the program and its partners (CSOs and C4Cs). • Rapid Response is more than a Roster. According to DAI staff and NIWETU stakeholders interviewed, 84 percent (11 male and seven female) felt that rapid response is more than a roster and is also about empowering local staff to innovate and leverage partnerships and networks. For Rapid Response Activities, NIWETU compiled a roster of independent consultants and issued agreements with 30 experts in the areas of (1) Interfaith Dialogue, (2) Government-Community Mediation, (3) Psychosocial Support, (4) Strategic Communications, and (5) Youth Counseling. The Rapid-Response Roster of NIWETU contributed to both mitigation and response, but there was a missed opportunity in linking this concept to Objective 2 and encouraging more effective field-based response systems. • Despite limitations and delays, the project contributed to some interventions that laid important groundwork for future CVE efforts. NIWETU, through its research and learning partner Wasafiri, issued seven TOs to Wasafiri, including: (1) the CAP Cycle Framework; (2) detailed 13 mapping of county-level VE actors and opportunities; (3) development of the scope and budget for NIWETU’s Baseline Assessment; (4) the subsequent delivery of the Baseline Assessment; (5) a study on VE recruitment dynamics in Isiolo County; (6) facilitation of a partner learning session; and (7) a rapid assessment of VE-related dynamics in the Boni Forest Reserve Area. These learning efforts provided fodder for adaptation and improved intervention design. 4.1.3 Evaluation Question 3 EQ 3: To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering VE and advancing its J2SR? What do key counterparts perceive to be the positive and negative aspects of the partnership? Definitions: “Extent,” defined as partnerships that have been scaled and have sustained the LOP as well as the post project period and are being carried out at the County or National level with leveraged local resources. Evidence There was mixed evidence that NIWETU’s systems approach was a feasible pursuit, given the range of factors beyond the control of the program that had an influence on partnerships. Interviews and documentation reviews showed that the “systems approach” was largely based on engaging a diverse set of stakeholders and targeting coordination and partnership, but it did not adapt or account for a delay in critical national and county partnerships in Year 1. This delay changed the phasing of activity interventions and created an over-emphasis on Objective 1 and a more minimal emphasis on Objective 2 needs and requirements. A more defined approach to systems, identifying which primary relationships impacted or impeded government capacity and community capability to work in the CVE landscape, may have benefited the project and created a more adaptive learning model of CVE in Kenya. While some stakeholder partnerships survived the life of the project, others were not able to sustain themselves—largely due to poor coordination or lack of feasible structures to ensure ongoing collaboration. Interviews indicated that NIWETU did contribute to improved partnership arrangements. Further evidence demonstrated improvements in relationships between government actors at multiple levels. For example, KSG/SMI has an established Memorandum of Understanding (MOU) with NCTC and the MoI—allowing them to collaborate on training and security issues relevant to border management and CVE. Because KSG is a state-run corporation that trains multiple ministries within the GoK, opportunities were noted in interviews about the role of NIWETU in establishing the Security Management Institute (SMI) and its sustained value as a legacy organization. Evidence shows that NIWETU attempted to establish private-sector partnerships as envisioned in the project design, but this did not sufficiently materialize or have the intended impact. This can be partly attributed to reticence on the part of the private sector to be associated with CVE and partly due to administrative delays in contracting. Evidence shows that there were also delays in establishing clear partnerships with NCTC and MoI. This lack of clarity about the roles of government stakeholders working on CVE affected capacity to address P/CVE in close coordination with the GoK during the initial year of the project. The success and strength of commitment, especially from GoK, were dependent on individual leaders’ personalities and priorities, with the consequence that the degree of commitment to drive sustainability of P/CVE efforts suffered in the early stages—impacting the time the project had to establish longevity and sustained initiatives post-NIWETU. “KSG and NIWETU has been one of the key partnerships created. The security Management Institute was created as a result. It is alive and vibrant. This has been meant to ensure that CVE is mainstreamed in government and civil servants have a point where they can get more CVE knowledge. The institute now has a Director, and a different program (CREATE) is funding it now. This collaboration has also seen the Ministry of Interior and Coordination of Government, NCTC and KSG even conduct training to the nonstate actors. There have been training modules developed for the different stakeholders like Office of the Deputy Public Prosecutor, Judiciary, and the National Police Service.” ~ Male KII, GoK Representative 14 Key Findings • There was mixed evidence on NIWETU’s systems approach. Based on the timing of GoK’s CVE strategy and a range of factors beyond the control of the program, partnerships were sometimes problematic. While some stakeholder partnerships survived the life of the project, others were not able to sustain themselves at the level envisioned in the project documents and in interviews with NIWETU senior staff. • There was minimal private-sector engagement. Despite evidence of effort to engage, there was minimal private-sector engagement despite this identified need. According to interviews and project documents, NIWETU attempted to establish private-sector partnerships as envisioned in the project design; but there was reticence on the part of the private sector to be associated with CVE. • Some NIWETU inputs continue to be used. Some of NIWETU’s interventions at the county level have been adopted and sustained beyond the life of the project. These include media products, CEFs, and the participation of C4Cs in security coordination meetings. Many of NIWETU’s civil society partnerships were adopted by other donor programmes working on P/CVE. • Weak government coordination had an impact. Based on interviews it was evident that weak government coordination had an impact on Objective 1 and Objective 2 outcomes and impacts. There was a lack of clarity about the roles of government stakeholders working on CVE, and this affected capacity to address P/CVE. There was also evidence of tension as to how this was handled between the field and Nairobi. This was particularly acute in the early stages of the project in partnership planning with NCTC. • CVE is multisectoral, and this was less of an emphasis in planning partnerships. According to stakeholders working in areas impacted by VE, CVE is multisectoral; and this was less of an NIWETU emphasis. NIWETU made limited deliberate efforts to delineate engagement aimed at security actors and civilian government actors. A multisectoral approach was needed to ensure more deliberate interventions for distinct state actors and local interventions. • Political will regarding CVE impacted NIWETU partnerships and implementation. According to field staff and partners, political will regarding CVE impacted NIWETU partnerships. The success and strength of commitment, especially from the Government, were dependent on individual leaders’ personalities and priorities, with the consequence that the degree of commitment to drive sustainability of P/CVE efforts suffered. 4.1.4 Evaluation Question 4 EQ 4: How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to VE? How can these interventions be improved? Definitions: “Effective,” defined as sufficiently designed to ensure that youth at risk for recruitment and radicalization were engaged or benefited from interventions. Evidence Evidence from interviews shows that throughout the life of the project, NIWETU made deliberate adaptations and adjustments to effectively include and work with at-risk youth. Based on field work, it was apparent, however, that interventions focused on a few sub-counties, especially in urban and peri-urban areas—thus neglecting youth in remote areas. There was evidence that youth were included in the development of some CAPs and contributed to the emergence of informal P/CVE networks, although the degree of influence varied by county. According to some FGDs and stakeholders, this participation mostly included more elite youth and did not address the need to include more diverse youth perspectives. It was also noted both by stakeholders and IP staff that most NIWETU interventions focused on prevention and not countering radicalization and “The design of NIWETU had the youth in mind. There were deliberate efforts to involve them from the very beginning. Apart from the planning that was for them, they were also the main beneficiaries of the project. They may not have been represented at the policy development stage, but they were also the majority running the CSOs which were implementing.” ~ Male KII, GoK Representative 15 recruitment. As this is branded as a USAID CVE project, a clear distinction between prevention and countering is critical in establishing activities and identifying critical GoK partnerships at the national and county levels. There was mixed evidence on how successful strategic communications and media products targeted at￾risk youth were. There was strong evidence that targeted messages were developed, addressing young people, but limited evidence on how effective the medium used was—as most youth have a diverse array of social media platforms that they access, and the project focused largely on YouTube shows. There was also an interruption in some of the planned media events surrounding this intervention, due to Covid-19 restrictions. Stakeholders interviewed also suggested that delays in contracting with media partners also impacted performance in this area. There were additional efforts made to improve strategic communications, but there were many suggestions made in FGDs and in stakeholder interviews to improve this component, particularly in impacting youth recruitment and radicalization risks. Some suggestions included digital training programs for youth, leveraging youth influencer twitter feeds, and capitalizing on youth events that focus on training in social media tools for youth issue promotion. Stakeholder interviews also revealed issues relevant to youth and CVE that were not able to be addressed within the framework of NIWETU interventions. Evidence from interviews indicate that due to government policy on returnees, the program was unable to offer support for returnees to integrate into the community—which proved to be an issue both of marginalization and of recidivism back into AS. The discreet nature of CVE programs does not support the use of communication platforms to amplify/raise awareness on existing returnee policies (Amnesty Policy by the government), and the project did not innovate to address this need. Aldo, previous experiences with returnee disappearances made their families wary of interacting with any amnesty offers and trusting government practices in ensuring integration back into community life. Key Findings • NIWETU identified and included at-risk youth in its interventions. Based on evidence, NIWETU identified and included at-risk youth in its interventions. These included training, media campaigns, and trauma healing. However, the definition of “at-risk” youth varied across activities and counties. According to interviews, there is a need to change the terminology of “at-risk” to address the changing nature of AS recruitment targets and address inclusion issues for elite youth capture of P/CVE activity interventions and grants. • Rural and remote youth were not as involved. According to stakeholders, rural and remote youth were not as involved. Interventions focused on a few sub-counties, especially in urban and peri￾urban areas, thus neglecting youth in remote areas. Over 80% (26 male and 13 female) of CSO and field staff interviews conducted stated this concern as an important consideration for future intervention planning. • Strategic communications interventions increased scope and reach. According to staff and partners, strategic communications interventions increased scope. There was strong evidence that targeted messages were developed, addressing at-risk youth, but limited evidence on how effective the medium used was. This was partially due to interruptions in planned activities due to COVID-19 restrictions. However, stakeholders also noted the changing and evolving types of social media platforms used by youth, AS, and communities seeking information on P/CVE issues. • Most NIWETU interventions focused on prevention. According to DAI staff and stakeholders, 86 percent (12 male and five female) stated that most NIWETU interventions focused on prevention and not countering radicalization and recruitment. Youth were included in the development of some CAPs and contributed to the emergence of informal P/CVE networks, although the degree of influence varied by county. Activities targeting reduced radicalization and recruitment were largely in the prevention space and focused on education. 16 • NIWETU youth interventions focused largely on education and not on mitigating AS radicalization and recruitment strategies. According to youth and stakeholders interviewed, NIWETU youth interventions focused largely on prevention education. Community-led initiatives largely focused on education, training, and other preventative activities. There was less of an emphasis on countering radicalization and recruitment, although media interventions and partnerships did address this to some extent. According to stakeholders interviewed, countering radicalization and recruitment is a targeted set of engagements that need to be better aligned with current and evolving AS recruitment methods. • There were unaddressed and dangling CVE youth issues. Based on interviews, there were CVE youth issues that were not identified until late in the project cycle. Due to GoK policy on returnees, the program was unable to offer support for returnees to integrate into the community. According to field staff and partners, NIWETU did not attempt to tackle controversial issues regarding this—such as confusion over amnesty policies that impact youth and families of returnees. Several interviews and FGDs noted this as a emerging and sustained issue relevant to development partners that needs to be addressed in a P/CVE activity. 4.1.5 Findings Specific to County Interventions This section summarizes findings from field-based interviews, which are triangulated with field observation and county-specific research. NIWETU project documents were reviewed as well as reports specific to each county, and their current and past profiles regarding VE incidents. The baseline, midline, and end-line were also used to confirm these findings and represent more specific county-level detail. Wajir County • NIWETU focused on counties and sub-counties, but a hyper-local focus was essential because AS had a hyper-local focus and took the time to build a grassroots recruitment/influencing strategy. This observation was made by multiple civil society partners. • NIWETU enhanced the role of women as frontline CVE actors/stakeholders. In the case of Wajir, which has a long history of women peacebuilders (the late Dekha Ibrahim, a globally renowned peacebuilding activist, was from Wajir). It seems it was an accidental/unintended impact, but NIWETU built on this long-standing tradition. In interviews, it was amazing that the women the ET spoke to owned this—and this is an opportunity to build on. In particular, NIWETU's training on early warning signs of radicalization for mothers and women had clear impact and could be expanded on. • There was not enough agility/flexibility in the NIWETU program in Wajir. At the very least, it did not keep up with AS's ability to frequently change tactics and approach. More agility and flexibility need to be built into future programs and accommodations made—if partnering with more cumbersome government systems. • AS targets social fault lines. As does not necessarily recruit based on ideological appeal; they target any kind of disenfranchisement. In Wajir, they have successfully recruited people with drug and substance abuse issues. There is no clear evidence so far that there was any kind of study or research around social networks and the strength of relationships/networks between different groups in society (women and youth, elders and youth, women and elders, and so on). Every social fault line is an avenue for recruitment and needs to be accounted for. • Because of the low development threshold in Wajir and multiple challenges in the environment which increase vulnerability to VE, it was necessary for NIWETU’s interventions to tie into other USAID/partner investments focused on job creation, economic growth, and skills training—to be able to offer young people opportunities beyond the awareness and capacity-building. • A frequent observation was how NIWETU managed to break the veil of silence around VE. Before the program, people were unwilling or hesitant to discuss this, and certainly not in public. Many feel that was the greatest achievement of the program: being able to break the silos and opening dialogue between different community groups. 17 • The lack of a conducive political environment was a particular change in Wajir. (There was a protracted legal battle between the Governor and his assembly, which led to an impeachment that was again fought in court and led to a reinstatement, which is also being contested.) This meant that the county was not as much of a vibrant partner even though they had an active role as envisioned in the CAP. • Wajir is a border county, and VE here has a cross-border dynamic—just like the society and economy, which require porous borders. It would be useful to pursue a cross-border approach and collaborate with USAID Somalia. Some peace-building programs did this in the past, but not NIWETU even though the hotspot areas were around the border. • While the national and county governments were collaborative to an extent, this sometimes did not trickle down to law enforcement. Police harassment is still persistent. There is a requirement for parents whose children are recruited to report to the police station regularly, and there also is no well-enforced amnesty program for potential defectors. • More investment in multimedia products and communication would amplify the reach of the interventions. Some investments were made, and it was clear that they had more sustained impact than the training and capacity-building efforts. Nairobi and Isiola Counties • Trauma healing continues to be an area that is poorly supported, with NIWETU largely pioneering this element in these two counties. Trauma among the police is also a factor to be considered, as it is possibly one reason behind some of the systemic negative behavior and mistrust between police forces and citizens. • While networks have been formed, and some sustained, others have fallen away due to the short duration of the grants. Across the board, the short duration of grants was mentioned as a key shortcoming in NIWETU’s design. • While CVE knowledge and skills have been enhanced, it is critical to match this with economic support to those vulnerable to recruitment, as access to livelihoods remains one of the key drivers of VE. • Entry points for VE actors are multi-directional, e.g., GBV, inter-clan conflict, election violence, corruption, etc. As such, CVE should also take on a broader approach to addressing the underlying issues that drive these other dynamics. • The brand “CVE” is not the most appropriate as it creates the “us” and “them” divide as well as a fear among some about being associated with it. For example, the private sector mentioned being quite averse to being associated with VE. Future USAID programs may want to consider branding P/CVE or coming up with other terms to mitigate these concerns, which hamper partnership. • In some cases, there was poor understanding among DAI staff about contracting and operating procedures for different kinds of actors. The boilerplate contracts applied to contracting was not fit for use for all partners, particularly the private sector or media firms and partnerships. • Many of the C4Cs engaged under NIWETU continue to be engaged in CVE and other community￾based activities. However, the CBOs formed are not operational because of their inability to access funding/win proposals—due to the stringent requirements for newer organizations. • Strategic communications around PVE is an area that should receive greater emphasis, with long-term support for it to be effective. This should be further embedded in government operations or partnerships designed to support these efforts and their sustained operations. Garissa County • Garissa was the site of the Garissa University College attack in 2015, which led to the death of 148 people, making it one of the worst terror attacks in Kenya. It also hosts one of the world’s largest refugee camps, the Dadaab Camp. Kenya’s official explanation for its military incursion into Somalia in 2011 was the frequent kidnapping of aid workers from this camp by AS. There is a long history of 18 communal conflict over land and other resources. Along with Wajir and Mandera, Garissa formed the Northeastern province, a historically marginalized and remote part of Kenya. • Disenfranchisement and historical grievances were manipulated by AS to drive recruitment into their ranks. Respondents spoke of a checkpoint on the outskirts of Garissa on the road to Nairobi, which effectively serves as border control, where residents must produce identification cards before proceeding “to Kenya.” Young people have difficulty accessing identification cards and are therefore effectively stranded and disconnected from opportunities. • Garissa sits on a thriving contraband route from Somalia. The GoK’s haphazard border closure strategy has inadvertently given rise to an organized criminal ecosystem where even otherwise legal products such as sugar must be smuggled in. Control over the smuggling routes and the potential to make money from this attracted AS to the territory. • NIWETU improved the relationship between security/government and the community. The improvement is primarily attributed to a shift in mindset and perceptions by the security apparatus as well as the community. This enabled information-sharing and collaboration. This is also credited for reducing the cases of extrajudicial killings and communal punishments that were associated with GoK in reaction to the terror attacks. • CVE interventions cannot be sustained or have long-term impact unless they are tied into other investments, especially around livelihoods. NIWETU may have addressed the drivers of VE, but the root causes and vulnerability factors can only be addressed through a holistic investment that includes employment, skills, and social amenities. This was particularly evident in interviews in Garissa and the historical trauma and impact of VE events. • VE dynamics in Garissa are changing, and the Boni National Reserve is the new epicenter. While NIWETU may have been successful at tackling dynamics in urban and peri-urban areas, Boni is the new hideout and staging ground for AS—and this is gradually undermining and reversing the gains made. • Clan and sub-clan dynamics factor heavily in Garissa's VE dynamics. Contestation over access and control of resources and formal power is manipulated by AS to garner sympathies or short-term support. • Just like Wajir, any P/CVE and peacebuilding investments in Garissa require a cross-border approach due to the strong social and economic linkages. • NIWETU identified and engaged vibrant civil society actors. Many of these actors have continued their work after the end of the program and have used the training materials developed by the program. • There was a limited investment in psychosocial support and trauma healing, despite consensus on both the need and potential impact. • The project implementation period was regarded as very short and inadequate by multiple stakeholders interviewed and in CSO grantee discussions. Mandera County • Mandera County is geopolitically and strategically important due to its position. It is referred to as the Mandera Triangle, where the borders of Kenya, Somalia and Ethiopia meet. The triangle has had a long history of conflict over scarce resources (pasture and water), communal conflict, and conflict over control of trade routes and markets. Since the collapse of the central government in Somalia, Mandera saw the influx of refugees across the border from Somalia. Notably, the then-deposed President, Mohamed Siad Barre, crossed into Kenya from Mandera before he was flown to Nairobi, Kenya. • Since 2012, Mandera has witnessed various attacks by Al-Shabaab that are ongoing. The attacks have predominantly focused on three targets: (1) non-local workers including teachers, quarry workers, and those in the service industry; (2) communication installations (telecommunication masts); and (3) public transport (particularly focused on Nairobi-bound buses thought to be transporting non-locals). In addition, there have been frequent attacks on security posts and government facilities. 19 • Due to the long, porous border as well as the vastness of the county, the government initiated a plan to build a border wall starting in 2015. This has not been completed and now remains abandoned. Also, the haphazard process in which the building was initiated was criticized by interviewees, because it did not consider cross-border trade and social networks. • In 2018, local authorities in Somalia protested that the border wall was encroaching on Somali territory—leading to a diplomatic spat. Notably, AS has used these claims to justify its continued attacks on Mandera. It has also justified its attacks on telecommunication masts by citing Kenya’s strikes against masts in Somalia. • In 2019, Cuban doctors working in a government hospital were kidnapped and moved into Somalia by AS. They remain in AS captivity. The government arrested their driver, a civil servant who was found guilty of aiding the kidnapping and sentenced to life in prison in 2022. • In Mandera, AS has manipulated long-standing inter- and intra-clan conflicts as well as grievances about the lack of opportunities for young people, poor education, and haphazard border closures by Kenyan authorities—which have impacted the cross-border trade that Mandera’s economy relies on. • AS attacks have led to the mass withdrawal of teachers, doctors, and other civil servants providing essential services, as well as quarry workers and service industry workers. This has devastated the economy and affected education and healthcare. These impacts require a multisystemic approach to the impacts and outcomes of VE. • AS has recruited heavily from Mandera and maintained an active presence in the town. In 2021, Mandera’s governor, Ali Roba, claimed that AS was in control of more than 60 percent of the county’s territory. The national government denied these claims but later stated that the group was significantly scaling up its operations in the county. • Mandera has a vibrant CSO landscape due to the long history of peacebuilding work within the county as well as across the borders with Somalia and Ethiopia. NIWETU worked with CSOs that were well￾regarded in the region, including Nomadic Assistance for Peace and Development (NAPAD), which has cross-border experience. • The county government was very supportive of the various P/CVE efforts in the county. However, multiple respondents mentioned that inter-clan tension hampered some of the goodwill. • Heavy-handed response by security forces was highlighted as a major challenge. Respondents stated that the communities in the county bore the brunt of violence from two fronts—AS and the Kenyan security forces. • Tensions between the Kenyan and Somali governments have exacerbated the country’s vulnerabilities. Kenya’s alleged support for a militia leader in Somalia saw the presence of these militia in Mandera and led to attacks into Mandera from Somalia. • Mandera is a perfect case study for why cross-border P/CVE programming is critical. 4.2 CONCLUSIONS 4.2.1 Evaluation Question 1: P/CVE Design and Theory of Change Application Effectiveness The design of NIWETU and the applied logic of the ToC was effective in assisting stakeholders to improve their CVE awareness and ability to identify VE threats. The ToC was ambitious, but the timing of some of the project interventions were more strategic than others—and the outcome of some of the interventions took time to take root and, therefore, to measure and assess their impact over time. The project largely used all quantitative indicators, with a research-and-learning subcontract to integrate adaptation and learning into intervention planning and phasing. The field team and IP management team often had varied opinions about the timing and scope of the interventions, but the observed result of the interventions was largely positive, and there was evidence that the project had contributed to outcomes that were still ongoing, had continued community ownership and, in some cases, been able to expand its geographic and demographic influence. 20 One of the major outcomes of the NIWETU design and activity was the way the project carved out space for CVE to be “desensitized” within community conversation and local drivers of VE to be more readily understood by a varied group of stakeholders. Interventions with women’s organizations were particularly impressive and had clear influence on the stakeholders and communities they engaged in. There was also strong evidence that these interventions were effective and could have been scaled up. NIWETU engaged a diverse set of stakeholders, including women and youth, and that participation was adapted based on research and learning to incorporate groups that were overlooked or underserved by project interventions. Important conclusions of NIWETU’s interventions under EQ1 are: • NIWETU created entry points for P/CVE partnerships in Kenya. The ToC and design of the project carved out space for CVE to be desensitized within community dialogue and a shared perspective on local drivers of VE to be more readily understood by a more varied group of stakeholders. The program’s learning and adaptation was not as effective in informing response, but the challenges encountered by NIWETU benefited multiple stakeholders and influenced other donor and GoK efforts on P/CVE. There were challenges in establishing clear linkages between national efforts and county-level interventions. But the lessons lived/learned and partnerships established have evolved P/CVE programming in Kenya. • Political will is essential in P/CVE design. NIWETU struggled early in the project to establish clear partnerships with NCTC and MoI, and this resulted in weak linkages between Objectives 1 and 2. Greater awareness by staff of governance issues evident at the county level, along with having clear partnerships with national-level entities, would have strengthened the ToC. Inconsistent government policy and lack of coordination within GoK was a challenge for NIWETU and is a sustained challenge for Kenya’s P/CVE implementation. • Citizen–government partnerships/coordination is essential. Partnerships and coordination could have been strengthened by a more systematic/dedicated approach within the program. Overemphasis on Objective 1 was an organizational as well as operational oversight. Despite initial momentum, some CSOs were not able to sustain their interventions due to security, influence, and resource constraints. The ToC did not adequately account for this vulnerability. • NIWETU’s ToC highlighted gaps in P/CVE design. Women are essential stakeholders and partners in P/CVE, not just victims or quotas to be included. Their perspectives should be centralized and not strategized. Gender plans and strategies are outdated for P/CVE, particularly in Kenya. “Youth at risk” is a term that is not relevant in P/CVE. Radicalization and recruitment impact communities and the emphasis on “youth at risk” is problematic. There should be a pivot to social inclusion principles and practices to mitigate the impacts of VE on communities. Formal P/CVE networks are essential to safeguard and ensure that civil society has a space where they lead the discussion. These networks can “invite” security elements but should not merge national security issues with human security needs. Challenges One of the weakest areas, based on interviews, was the interconnection between Objectives 1 and 2, specifically linking human security needs (county/sub-county perspectives) and national security partnerships. This was an important component area of the ToC and, as a result, outcomes—although largely positive—had marked areas of needed improvement in a P/CVE project. There was a delay in establishing effective relationships with GoK at multiple levels. This likely contributed to the challenges in establishing clear linkages between national efforts and county-level interventions. Inconsistent GoK policy and lack of coordination within GoK was also a challenge for NIWETU. There was evidence that partnerships and coordination could have been strengthened by a more systematic/dedicated approach within the program, particularly during Year 1. 21 There was limited evidence of the establishment and impact of formal networks, but there was evidence that NIWETU’s activities contributed to the emergence of informal networks that continue to engage on CVE. The CEFs were more of a monthly dialogue than a formal network. Networks were established in most counties that have survived beyond the life of the project, but the efficacy of these networks varied. The project design was unclear about the definition of “networks established.” This created flexibility in identifying impact but also may have been less impactful for a CVE project. NIWETU’s community-driven-and-led approach to CVE focused on community dynamics in each identified VE hotspot and engaged a diverse group of stakeholders and community elements. However, it was not always possible to engage certain key stakeholders based on community tensions, relationships, weak political will and, in some cases political paralysis. According to senior DAI staff and USAID counterparts interviewed, the ToC was effective in identifying the community elements most impacted by the radicalization process, but some elements were negatively impacted by the timing of specific partnerships (i.e., delays with NCTC, MoI, etc.). There were also challenges with targeting and the evolving nature of VE hotspots and locations at the county level. Opportunities There was evidence that NIWETU’s interventions under Objective 2 contributed to improved government understanding and capacity to deal with VE. There are opportunities to build on this in subsequent programming. Interventions were intimate: designed to impact the individual, family, and community based on specific conditions and circumstances of the targeted counties. However, it was noted that VE is changing in Kenya, and there is a constant need for ongoing analysis and outreach at the national and sub-county levels, particularly. NIWETU established the KSG/SMI to assist in better analysis and training for government officials working on CVE. The KSG/SMI intervention has proved a remarkable legacy for future P/CVE work in multiple sectors impacted by security, cross border management, or human security issues. KSG/SMI’s potential as a future partner and for other donors should not be overlooked. Improvements Although the ToC for NIWETU managed to evoke all the anticipated result areas, future designs would benefit from a more simplified ToC that looks for direct linkages between national efforts on CVE and interventions at the sub-national level. It is essential for future P/CVE efforts that seek to strengthen the implementation of national or sub-national CVE policy partners and link to potential localized funding sources and/or development planning processes at the county and sub-county levels. The timing of NIWETU and subsequent efforts opened the opportunity for improved CVE programming with other sub-national and donor efforts planned and ongoing (i.e., CREATE). Future P/CVE efforts should focus on localized approaches that could integrate more seamlessly into local development planning and funding allocation (i.e., education efforts, mental health, and job creation or economic growth/livelihood programs). Future programs would benefit from focusing on “legacy” interventions at the national and community levels—like KSG/SMI—that are not bound by time constraints. Time needs to be approached differently within a CVE program. There is a tension with a “quality vs. quantity” dynamic—working with fewer, established entities as opposed to working with a wide array of less-established local actors. This would change the way that subgrants are designed and administered (not Office of Transition Initiatives [OTI], not capacity-building, but a hybrid approach). Local civil-society actors need more exposure and “The relationship between the community and the police is very fragile. One incident could ruin the trust which has been built. We believe we have enhanced the relationship between the community and the police. However, two years ago the police killed an innocent person, and the community was very angered. The relationship that has been building disconnected. The police and the community spent a few days in running battles where 5 people died and scores were injured. There is need for continuous police and community engagement. They are also traumatized”. – Female KII, CSO Representative 22 interaction with national security architecture. There is sustained and well documented mistrust between civil society and security actors. There is a “reverse gap” in CVE response awareness that needs to be backfilled with a more relevant and current understanding of national security and its roles and responsibilities at the county, sub-county, and cross-border levels. A new program could have a cross-border approach through a direct partnership with another program (in Somalia). A sub-national cross￾border approach is also essential for local borders (between counties) and could be leveraged either through contract requirement or design element (i.e., strategic communications, network partnership, or civil servant exchange activities). More flexibility/agility should be a deliberate strategy of the program to tackle/address changing trends and counteract AS dynamics. More elements that focus on rapid response, deep networking/information sharing, or reporting mechanisms for risk are essential in P/CVE programs that seek a link between national security and the sub-national (county/sub-county) levels. A prescribed governance intervention is necessary for an effective CVE program. This could be through a liaison entity (such as KSG) or a clear design that addresses issues of governance essential to improving human security issues—as opposed to implementing CVE or national security policies. There is a need to understand how devolution affects human security and CVE. How has devolution affected local security? What security roles can be further devolved? Understanding the opportunities and challenges presented by devolution are also essential for future P/CVE efforts. 4.2.2 Evaluation Question 2: P/CVE Interventions at the Hyper-Local Level One of the most effective interventions was subgrants to increase the scope of diverse actors exposed and engaged to P/CVE. Subgrants created space for community-led initiatives, enhanced ownership, and strengthened diversity of stakeholders. In Year 1, USAID approved 17 grants, five of which were completed with local partners. Initial subgrants focused on grants to CBOs to address locally relevant drivers of VE, a Strategic Communications Pilot Project, and Rapid Response Activities. Subgrants are an important tool but can also limit innovation when seeking more sustained technical partnerships that require a contracting mechanism. Future interventions would benefit from a more flexible set of tools for funding partnership initiatives. Despite initial challenges with national partnerships, NIWETU contributed to other forms of coordination that were helpful for USAID and partners. NIWETU also established monthly CVE Program Coordination Meetings for CVE programs in Kenya, which brought together participants from USAID and other donor￾funded programs. As a Strategic Communications Pilot Project, NIWETU worked with USAID to review and provide feedback to United States International University–Africa (USIU-A) on the Strategic Communications Pilot Plan. NIWETU was very successful and strategic at selecting the right persons to serve as C4Cs. The selection process was participatory and contributed to increased cooperation between community stakeholders. C4C network members were effective influencers—recognized by the GoK, communities, and CSOs. C4Cs have continued their work even after the end of the project, acting as community-government liaisons and providing psychosocial support and advocacy support for CSOs. C4Cs were able to amplify key messages developed in the strategic communications efforts and were integral to conducting training in P/CVE for the GoK, communities, and schools. They improved the learning and adaptation process of the program and continue to support the establishment of informal networks that create space for community identification and efforts to mitigate VE impacts at the local level. Improvements Some of the improvements necessary for future efforts where networks are utilized as a tool to increase scope are that individual C4C need to be better resourced. One of the ways that the project attempted “Future projects should focus more on grassroots and cross-border areas since counties in Northern Kenya border Somalia and AS groups exploit the porosity of these borders to recruit and radicalize cross-border communities.” ~ Male KII, GoK 23 to create longevity of these networks was to encourage the establishment of CBOs. This was an unnecessary investment since the individual profiles were more visible and impactful and the establishment of CBOs required greater assistance to be fully established as viable organizations. Individuals were themselves powerful influencers and their informal networks in many cases were established enough to spread PVE and CVE messages in their communities. One of the interventions that proved valuable for communities impacted by VE was psychosocial healing. These interventions were recognized by communities and other stakeholders as necessary for strengthening the community’s resilience against VE. Improvements are necessary, however, to ensure the use of locally available facilitators/counsellors to make it more impactful. These sessions, which have been identified as crucial for addressing historical grievances that act as drivers for VE, can also provide a valuable service to security professionals working in local communities. Extending trauma healing and psychosocial support to police and other security agencies can improve the relationship between the community and security agencies. Significant mistrust between community members and national security institutions at the county level was a challenge in several of the counties in which NIWETU worked. Several CSO members interviewed suggested that this is an area that may have a natural and reasonable fault line that needs to be respected, not blurred. Delays in Year 1 and the significant reluctance of key partners (MoI and NCTC) to engage in early stages of the program impacted initial interventions. As stated by numerous stakeholders, GoK and community willingness to engage with NIWETU activities and work together to address VE in hotspot counties and at the national level, particularly with key national-level partners (i.e., MoI, NCTC, and KSG/SMI), was critical to success. Government stakeholders and USAID staff noted that high-level GoK interlocutors were important in creating space for partnership with NIWETU—and these need to be cultivated in advance of startup, if possible. Key conclusions of NIWETU’s interventions under EQ2 are: • NIWETU discerned the “P” from the “C” in P/CVE. Objective 1 was dominant in the NIWETU landscape. Three of the four result areas were addressed in this space: 1) improved CVE knowledge and skills; 2) community networks established; and 3) community-led strategies and initiatives. NIWETU’s research and learning was broad, ambitious, and useful for future programs; but was too complex in most interventions to be integrated and sustained with local partners and efforts. In P/CVE programs, multiple mechanisms—beyond grants—are needed for funding initiatives. More responsive and flexible funding is needed for emerging priorities and needs. Subgrants improved CVE knowledge and skills, contributed to informal networks, and created space for community-led strategies and initiatives. • Organization capacity-building is unique for P/CVE. NIWETU worked with many CSOs that have systemic mistrust of security institutions and have designed interventions to mitigate and to collaborate with these issues. CSOs working on P/CVE issues need more education on the national security architecture in Kenya for VE response. Human security and national security are distinct, and NIWETU partners highlighted some of the essential elements necessary to ensure that human security becomes more of a priority for county-level partners working on CVE. • NIWETU interventions that increased scope. CVE Champions increased the reach of NIWETU and worked beyond the sub-county level. C4Cs were effective influencers—recognized by the GoK, community, and CSOs. Their hyper-local contributions improved the learning and adaptation process of the program within the counties. Community-led interventions such as trauma healing have a significant impact on both PVE and CVE. Trauma healing provided a safe space for affected community members to address the impact of VE and their vulnerability at both the individual and community levels. Strategic communications have the potential to catalyze greater impact beyond the constraints of resource and geographic scope. NIWETU’s activities 24 emphasized the communication and not the strategy. There is a need to adjust these interventions to match the current digital landscape in Kenya. 4.2.3 Evaluation Question 3: Partnerships, Resilience, and J2SR In addition to its frequent engagement with government officials at the county and national levels, NIWETU succeeded in partnering with the GoK on two major initiatives: 1) supporting the development and launch of CVE County Action Plans (CAPs), and 2) the founding of SMI at KSG. Partnerships NCTC Partnership: Supporting CVE County Action Plans (CAPs) NIWETU’s support for 41 RCAPs was one of its signature achievements. When NIWETU launched in 2016, there was little county-level implementation of Kenya’s National Strategy to Counter Violent Extremism (NSCVE). While NSCVE provided a framework for CVE work, its nine pillars8 had not been operationalized on the county or local levels. With support from donors, GoK launched the so-called “first wave” CAPs in four of Kenya’s five coastal counties in 2017 and 2018. However, except for the Mombasa CAP, these plans saw limited success due to lack of political will and financial and institutional support for CAP implementation. In 2018, NIWETU began working with national, county, and local CVE stakeholders to support development of CAPs in four of its core counties: Garissa, Isiolo, Mandera, and Wajir. Each five-year CAP defined the unique VE context of the county and prioritized which of the NSCVE’s nine pillars should be tackled first to address the VE threat. Drawing on the expertise of a diverse group of CVE stakeholders in each county—including youth, women, religious leaders, government and security officials, and media representatives—each CAP also included additional pillars tailored to the specific VE challenges in that county. For example, stakeholders in Isiolo added a “Women Pillar” to recognize the important role that women and girls play in both VE and CVE activities and to ensure that women-focused CVE programming was a CAP priority from the start. After the CAPs launched, each county founded a CVE CEF tasked with implementing the CAP and coordinating CVE activities across the county. Each CEF is co-chaired by the governor and county commissioner and comprises government officials, security officers, and civil society representatives. New CVE stakeholders and activities in the county must work with the CEF to ensure that their activities are contributing to CAP implementation and correspond with the county’s priority pillars. Observations for Improvement NIWETU saw mixed success in CEF engagement and CAP implementation progress. As one might expect, CEFs were most successful when local national and county government officials showed a strong interest in CVE and prioritized CAP implementation. In Isiolo, the CC and County Secretary both showed early interest in CAP implementation. The CC in particular championed CAP implementation and ensured that the CEF met monthly. The consistency of CEF meetings and strong engagement from the CC secured early CAP implementation successes in Isiolo that was lacking in other counties. On the other end of the spectrum, county and national officials showed little interest in CAP implementation in Wajir, where progress in CAP implementation was slow despite steady pressure from NIWETU and CSO partner staff. After completing the initial four CAPs, NIWETU and its government and CSO partners had the opportunity to reflect on what worked—and what could have been improved—in CAP development. This research proved invaluable when NCTC asked NIWETU to help it to develop 37 RCAPs in all remaining Kenyan counties. AS’s attack on the 14 Riverside complex in January 2019 laid bare something that NIWETU and its partners had foreseen in the years leading up to the attack—that VE groups were making inroads into new 8 Pillars include Psychological, Education. Political, Security, Faith-Based and Ideological, Training and Capacity Building, Arts and Culture, Legal and Policy, and Media and Online. 25 communities, and recruitment was no longer restricted to the “hotspot” counties on the coast and along the Somali border. In the wake of the attack, President Uhuru Kenyatta issued a Presidential Directive for every county to develop a CAP before June 30, 2019. This request from NCTC spoke to the close relationship that NIWETU had developed with its GoK partners that allowed NCTC to approach NIWETU to help it accomplish this monumental task. In each of the remaining 37 counties, the five-month CAP development process was distilled into three days of intensive engagement with security, government, and civil society stakeholders in each county. NIWETU staff accompanied its partners NCTC and Malaika Foundation (MF)—a Kenyan CSO that had supported the four previous CAPs—during an exhausting two-month flurry of RCAP development. Despite the complexity of this activity, RCAP development largely went off without a hitch due to the camaraderie between NIWETU and its partners, the flexibility of NIWETU’s grants and logistics team, and the successful operationalization of learnings from the “second wave” of CAPs. Kenya School of Government (KSG) – Security Management Institute (SMI) Partnership Through constant engagement with its government partners, NIWETU successfully supported the founding of the SMI at KSG. The SMI is the first CVE government training and research hub in Kenya and, indeed, is the first CVE-focused think tank in East Africa, presenting an opportunity for KSG to lead the regional academic and policy conversation. SMI continues to provide CVE trainings and official certificates to high-level GoK public servants, while simultaneously mainstreaming CVE knowledge and skills into all training programs. KSG has signed MOUs with NCTC and MoI to lay the groundwork for sustained and future trainings and joint CVE research assignments. In a GoK environment defined by inter-departmental rivalries—especially between organizations with overlapping mandates, these sorts of partnerships would not have been possible without consistent NIWETU engagement and mediation with GoK stakeholders. Observations for Improvement The level of trust and mutual respect between NIWETU and its GoK partners that was required for both the RCAP and the SMI activities to succeed should not be understated. Neither of these activities would have been possible without years of consistent engagement between NIWETU staff and key GoK officials, including the NCTC Head of Prevention and Resilience, Njenga Miiri, and the MoI Deputy Minister for Internal Security, Thomas Sakah. By showing support for GoK-driven initiatives like the CAPs, NIWETU was able to demonstrate its willingness to work with NCTC in good faith on future activities, including the SMI. By the time the SMI officially launched in 2020, NIWETU staff members were communicating with Miiri and other top officials daily. NCTC trusted NIWETU and actively sought NIWETU advice, and NIWETU honored that trust by encouraging GoK agencies to take credit for their achievements. However, delays in the early part of the project undermined the ability of NIWETU to effectively leverage this relationship at the county level. Opportunities were missed that the project would have benefited from. Some other important conclusions of NIWETU’s interventions under EQ3 are: • NIWETU’s systems approach was a partial success. The systems approach only addresses a partial set of issues relevant to implementing a national P/CVE strategy. To the extent possible, NIWETU contributed to advancing the J2SR principles. Multi-stakeholder involvement in the development of CAPS and establishment of CEFs established and deepened relationships and “Future partnerships need to understand the context under which they are operating. NIWETU had a partnership with HEDAYA which did not work out. The company was from Dubai and did not understand the local context. There was no consultation with the government and other key stakeholders. It would have been a good document. People from outside cannot come and tell us what to do. We understand VE in our country better than they do. We have used this learning to advise other programs (CREATE) against taking the same route. We have told them that while CSOs are doing a great job in CVE, they cannot exist in a vacuum. They must be guided by NCTC. Programming on CVE needs to be anchored on the national strategy with NCTC playing an oversight role. Community stakeholders only need to operationalize what has been approved by NCTC.” – Male KII, NCTC Staff 26 contributed to enhanced capacity for preventing and countering VE. NIWETU diversified the landscape of partners working on CVE. Multi-stakeholder involvement in the development of CAPS and establishment of the KSG/SMI deepened relationships and contributed to enhanced capacity for preventing and countering VE. • P/CVE programs need a multi-sectoral approach to be self-reliant. Private sector partnerships can address socioeconomic fault lines as well as PVE-relevant inclusion issues. There was no clear engagement modality for private-sector partnerships. This was a missed opportunity to innovate a role for the PS in P/CVE as well as leverage local resources in support of J2SR. NIWETU identified CVE gap areas that require the commitment of GoK across ministries and service sectors. Some NIWETU interventions helped identify gaps in services as well as the commitment necessary to adopt P/CVE tools and interventions (CEFs, psychosocial needs, returnees, amnesty issues, etc.). There is a fault line between human and national security concerns that AS capitalizes on. Jurisdictional issues hampered collaboration, commitment, and adoption of NIWETU tools. USAID’s P/CVE programs need to draw clear lines between security interventions and governance interventions to maximize impact and mitigate risk (CVE toolkit/ NCTC). 4.2.4 Evaluation Question 4: P/CVE Targeting, Inclusion, and Marginalization NIWETU interventions had a strong focus on targeting at-risk youth (male and female). However, narrowing the program’s focus to at-risk youth, especially from underprivileged backgrounds, was non￾inclusive and created a loophole for VE groups to target youth not identified as at risk. There is a need to switch emphasis from efforts targeting “at-risk youth” to designing activities that provide opportunities to enhance inclusion and social cohesion and address vulnerability factors. Future P/CVE programs should focus on strengthening social cohesion and issues of exclusion evident from changing patterns in AS recruitment that target coaches and teachers as portals for recruiting youth. More established partnerships with youth associations may broaden inclusion and minimize elite capture of youth programming by more established urban youth groups. NIWETU’s flexibility in adapting interventions to the changing dynamics on the ground supported the program in achieving its objective for reaching a wide variety of at-risk youth, although there was a gap in adequately reaching youth in remote and rural areas. Linking CVE interventions to existing GoK initiatives such as the Youth Enterprise Fund and CDF funds, as well as other donors (World Food Programme [WFP], World Bank), would improve donor impact. Improved financial inclusion for youth to access government funds by enabling sharia-compliant financial products would also be more relevant for Muslim youth and create greater inclusion. Observations for Improvement Although NIWETU’s interventions used various communication platforms in CVE interventions, effectively reaching youth requires an understanding of how youth consume and interact with information. There is a need to increase innovation in strategic communications to leverage newer media and messaging techniques (TikTok, Twitter spaces, etc.). Evidence shows that while the program worked to engage youth in P/CVE activities and dialogue, previous and ongoing experiences of youth with enforced disappearances and police brutality is an issue that still requires continuous engagement. The discreet nature of CVE programs does not support the use of communication platforms to amplify/raise awareness on existing returnee policies (e.g., the Amnesty Policy by GoK). Also, previous experiences with returnee disappearances make families of returnees wary of interacting with any amnesty offers. The issue of the status of returnees remains a challenge with CVE in “Women and youth should be targeted and integrated even more during these projects because they make up a large number of the population; and women hold significant sway in the community. Also, youth are not homogeneous and targeting them should be with that lens of recognizing their differences and reaching them appropriately.” ~ Female KII, Youth C4CF 27 Kenya and for affected families. There is a need for collaborative efforts between government and CVE actors to agree on the modalities of supporting the reintegration of returnees. Important conclusions of NIWETU’s interventions under EQ4 are: • The term “at-risk” can undermine P/CVE programs working on inclusion. NIWETU’s targeting of at-risk youth (male and female) was not always aligned with P/CVE needs at the county and national levels. There was a limited focus on youth issues as they relate to CVE contexts (i.e., reintegration, deradicalization, counter-messaging, etc.). The targeting of youth at risk of radicalization and recruitment as an element of a P/CVE program was more effectively done by local CSOs, which had a better understanding of the context and risk factors. The issue of returnee reintegration is an area of CVE that does not yet have a “home” in USAID P/CVE efforts. Programs that focus only on “at risk” sometimes neglect opportunities to assist in other critical effects of VE, such as the reintegration of returnees. • Recruitment and radicalization are a community concern. There is increasing evidence that to address radicalization and recruitment of youth, programs should also focus on women￾led initiatives seeking to mitigate risks. These efforts tend to be multi-sectoral to address drivers as well as risks. Recruitment and radicalization are urban and rural issues alike. To address issues of geographic reach, NIWETU worked with the C4Cs, expanding to secondary locations within counties, and used media partnerships to expand messaging. Although NIWETU’s interventions used various communication platforms in CVE interventions, effectively reaching youth requires an understanding of how youth consume and interact with information. It also requires including impacted communities and families to design relevant content. 4.3 RECOMMENDATIONS 4.3.1 Recommendations for P/CVE Design (Relevant to EQ1) Recommendation 1: Design future P/CVE programs with a prescribed governance intervention as well as a clear partner within the National Government. CVE programs require extensive collaboration at the national level. Donors and development actors require clear GoK partnerships to implement P/CVE interventions at the sub-national level. Recommendation 2: Account for evolving devolution impacts in new P/CVE programs. There is a need for P/CVE programs to investigate how devolution affects human security and CVE. Devolution is affecting local security as well as the roles and responsibilities of these institutions. Understanding the opportunities and challenges presented by devolution is crucial to the success of future P/CVE interventions. Recommendation 3: Future P/CVE programs should be multisectoral. P/CVE programs should integrate more effectively with other sectoral projects and not only focus on hyper-local conditions. VE is a long-term issue. P/CVE activities should be designed to interlink with other development efforts and funding (i.e., education efforts, mental health, and job creation or economic growth/livelihood programs). 4.3.2 Recommendations for Interventions and Learning (Relevant to EQ2) Recommendation 1: Grant programs are a critical tool but are also time-consuming. Fewer grants with established partners that have the capacity to prime and include smaller organizations in their work planning may leverage better outcomes. Prioritizing strategic partnerships with well-established organizations could yield better impact over time at the community level and contribute to advancing J2SR. This may also allow for an emphasis on multi-sectoral opportunities relevant to CVE. Recommendation 2: Future programs need CVE champions and influencers. Citizen-centric engagements are an effective modality for enhancing community capabilities around P/CVE. Future programming should scale up and dedicate more resources to C4Cs or a similar mechanism. 28 Recommendation 3: Strategic communication is essential in future P/CVE programs and can also address gender gaps. These interventions need to mature to address the changing social media and digital landscape and to strengthen the voices, perspectives, and leadership of women impacted by VE. Future P/CVE projects should have dedicated funding streams to leverage strategic communication efforts and integrate them into field-based activities that aim to enhance reach and demographic scope. Recommendation 4: Trauma healing is an essential intervention to enhance social cohesion. It is necessary to expand psychosocial support interventions to police and other security agencies to support the security organs’ resilience and capacity to respond to VE. 4.3.3 Recommendations for Resilience and J2SR (Relevant to EQ3) Recommendation 1: Future P/CVE programs should focus on leveraging “legacy” interventions that have sustained benefit to GoK. The relationship with KSG/SMI is an asset that should be used in future CVE programs. Recommendation 2: Future programs should address self-reliance early and high, as VE is a long-term issue in Kenya. P/CVE interventions and investments should be mainstreamed into the development agenda at the county and national levels. This would make a systems approach more impactful by emphasizing ownership over partnership. It would also allow economic realities to be aligned with interventions. Recommendation 3: VE is a private-sector issue, and the private sector is a critical partner for P/CVE programming. Develop an improved private-sector engagement strategy specific to P/CVE that creates entry points, addresses barriers to partnership, and prioritizes partners that may be directly affected by VE (Safaricom). Interventions should emphasize areas of excellence such as strategic communications, media, and skills training. It is also necessary to dedicate resources such as specialist technical expertise within the program. Recommendation 4: USAID should maintain key leadership engagements with multiple government stakeholders even when there are no active P/CVE programs. The goodwill and momentum developed in NIWETU can be leveraged by a future program to leapfrog existing systemic issues and facilitate a degree of co-creation for future interventions. 4.3.4 Recommendations for Future P/CVE Targeting (Relevant to EQ4) Recommendation 1: Programs with flexible funding may bridge inclusion gaps. Explore explore more flexible funding and engagement that allows for more agility and adaptation to respond to changing VE dynamics and exclusion issues and is more responsive to emerging trends and issues that arise. Recommendation 2: Future programs should focus on risk factors and not “at-risk” youth. Switch the emphasis from efforts targeting “at-risk youth” to designing activities that provide opportunities that enhance inclusion and social cohesion and address dynamic vulnerability factors. Recommendation 3: Innovate to impact youth vulnerability and VE impacts. Have a clear focus on P/CVE with a systematic criterion to identify target youth across demographics and geographic locations. Do not limit partnerships to youth but include women-led organizations focusing on VE impacts. Recommendation 4: Exclusion and social protection needs are critical vulnerabilities leveraged by Violent Extremism Organizations (VEOs). Think through the risk and vulnerability of youth differently—and focus on disenfranchisement as opposed to emphasizing abstract socioeconomic and political factors when targeting stakeholders. Recommendation 5: Expand digital space for messaging and diversify voices. Make room for innovation in strategic communications to leverage newer media and messaging techniques. Include women and integrated perspectives on VE impacts on youth and their communities. 29 Recommendation 6: Returnees, Radicalization, and Recruitment are interwoven in Kenya. Future programming should be linked with other initiatives working on Disarmament, Demobilization, and Reintegration (DDR). While this is highly dependent on political will, a future program can act as a facilitator for the community component of DDR. 4.3.4 Recommendations for Future P/CVE Programs Recommendation 1: A future P/CVE program should incorporate a liaison officer seconded to the county governments as well as with NCTC. The liaison officer at the counties would be responsible for improving the relationship and political buy-in from the county leadership as well as facilitating relationships for field and program staff. The liaison officer would have deep insight into both the program’s priorities as well as the county administration’s priorities and contribute this knowledge to the design and implementation of realistic, impactful interventions. These positions can be filled through a joint recruitment by the program and GoK. Challenges with this approach may include the possible pay disparities between the liaison officer and counterpart civil servants as well as ensuring adherence to USAID and GoK HR guidelines. Recommendation 2: Regular political economy analysis is necessary for the success of any P/CVE programming. This is especially pertinent in NIWETU’s focus counties because of the deeply embedded governance, political, and security issues that drive P/CVE beyond extremist ideology. Political economy analysis could happen at three levels: a. Mission-Level: Through analysis of open-source information, including media monitoring, key leaders engage with government, civil society, and periodic data collection (focus groups, perception surveys). b. Program Level: At this level, the program could focus on hyper-local trends and developments—including conflict dynamics, social and economic trends, and the impact of specific attacks and methods of operation by AS in their radicalization and other operations. c. Activity Level: Informed by the above two, each intervention should incorporate actionable political economy analysis. This will be owned by the field staff and replace the complexity awareness mechanism that was used by NIWETU with limited success. Recommendation 3: Devolve greater authority to the field. Because the bulk of implementation takes place in the field, decision-making should also be devolved/centered in the field. The Nairobi team could play a secretariat role, but the bulk of the grant-making team and community engagement activities/decision-making should be in the field. Recommendation 4: Implement multisectoral programming by co-implementing with other USAID programs, staff, and initiatives. Improving linkages at the field level with other USAID programs and their staff is also crucial. This not only contributes to improved analysis and understanding of the political economy but also provides to cross-leverage opportunities for learning, co-design, and co￾investments. 4.3.5 Improving CVE Programming in Development The current USAID Policy for Countering Violent Extremism in Development Assistance9 does not include a clear delineation between programs that address VE drivers, those that reduce risk, and those that seek to build CVE capability in the countries in which USAID works. A clear understanding and description of those differences is important for the mission to understand how both VE (the problem) and countering VE (the means to solve the problem) might be assessed and addressed in development planning and intervention design. There are essentially two categories of CVE programming: (1) VE Risk Reduction Programming and (2) CVE Systems Programming. 9 USAID Policy for Countering Violent Extremism in Development Assistance, USAID April 2020. 30 Risk Reduction Programming addresses the reasons that people find their way to radicalization and VE. These programs address the drivers, particularly the push and pull factors10, which are context-specific and vary by country and location. These are the factors that reduce the likelihood or risk that individuals might convert to VE or engage with VEOs to recruit or radicalize vulnerable populations. As we have seen in the evolution of AS’s tactics to radicalize and recruit, they no longer merely seek youth but also teachers, coaches, and influential members of communities in which they operate. They often seek secondary influencers to recruit youth and other members into their movement. In Application, Risk Reduction Programs: • Currently account for most of the programming targeting VE within USAID (central and mission programs). Most of these programs are PVE and not CVE programs, despite the classification in the contract or in USAID branding. • Address the context-specific drivers of VE and have end goals that reduce the presence of these drivers in targeted geographical areas. These programs will always require hyper-local approaches with enhanced assessment and PEA. • These programs emphasize targeting people identified to be at risk of radicalization and recruitment. These targets are critical in achieving success and often are under-targeted by not emphasizing the role of women and community influencers in the recruitment and radicalization strategies of VEOs. • USAID and development programs in this category at best are “CVE sensitive” and are primarily focused on prevention, or PVE. There are often weak reporting and measurement methods for tracking reduction in radicalization or decreases in recruitment. • Most programs also focus on reducing risk but fail to identify ways to articulate changes in the context that can be directly attributed to development program interventions. There is a need in these programs to have an independent body work alongside the IP to conduct baselines, midlines, and end lines to assess, track, and document contextual and behavioral change to establish risk reduction in factors that lead to engagement with VEOs. CVE Systems Programming builds a nation and community’s ability to recognize and intervene to prevent and counter VE. These programs do not address drivers; instead, these programs work to create CVE policies, strategies, and institutions in partner governments. They seek to train and prepare government officials to recognize and understand and work with citizens and local communities to prevent VE. These programs work at the national, sub-national, and local level. Populations targeted are inclusive and include CSOs, teachers, religious leaders, special interest groups, the private sector, parent-teacher associations, community leaders, and advocacy groups—who are trained to not only recognize the signs of VE but also to have the agency, authority, and power to intervene in the process. In application, CVE Systems Programming: • Accounts for a very small fraction of current USAID programming. NIWETU was largely a PVE program with some elements of CVE Systems Programming. • These programs do not address the drivers of VE; instead, programming seek to empower and activate national, sub-national, and local governments to empower communities to engage and lead CVE efforts. • These programs work on strengthening institutional capacity and community capabilities to address VE by creating systems to cooperate, partner, and collaborate on critical intervention needs. NIWETU did not require a “systems approach” but, rather, a “systems way of thinking” about CVE needs and targets. • CVE systems programs do not seek to target at-risk populations but, instead, improve safety nets and social protection systems and address systemic grievances and critical needs to strengthen community and sub-national response to VE and VEOs. These programs may seek to increase access to livelihoods 10 The Development Response to Violent Extremism and Insurgency, USAID 2011. 31 and critical services to reduce the vulnerability of communities to engage and support VEOs seeking to operate in their locales. • These programs intentionally seek to build frameworks, structures, and systems that can operate with local resources (social capital and funding) and directly engage current budget and organization constraints of national and sub-national partner institutions. KSG/SMI was a “legacy intervention” of NIWETU and reflects a CVE systems approach to CVE programming. This is one of the few interventions within NIWETU that fall into this category. • CVE systems programming requires national collaboration through established MOUs or partnership with critical national or sub-national structures. The timing of NIWETU was appropriate as Kenya had a new National CVE Policy, but some of the interventions failed due to weak ownership and co-creation with NCTC and MoI. When the project attempted to have partners adopt a CVE toolkit, it was rejected by NCTC as it was felt the process did not reflect the needs of NCTC and the national CVE landscape. As the toolkit process was largely done with local partners (CSOs and stakeholders), it was oriented toward civil society and not toward the shared spaces of partnership and implementation that were necessary at the sub￾national level to implement the national CVE policy. • CVE systems programming requires working with elements of civil society that engage or need to engage on national security and human security issues. There is a grey space that exists in CVE that requires development partnership, because the intervention’s end goal is to improve community security and well-being. Countering VE requires community support and sustained trust between citizens and national security institutions. There is a critical need to educate civil society and constituent populations about the role of security institutions. There is a critical need to strengthen the role of citizen engagement in overseeing sub-national security institutions to ensure that this becomes an enduring role. “The CVE toolkit is an intervention that did not work. As much as VE is VE wherever you go, there was lack of contextualization during the development. There was a feeling among different stakeholders that the toolkit did not speak to the Kenyan context. In future, an external organization like HEDAYA who were leading in the development of the toolkit could review the document but not design it. We saw interventions that were led by local organizations having better understanding of the local issues. A good example is what happened in Kamukunji where KACPEN researched on the VE context, understand the context, and came up with an action plan relevant to the community.” ~ Female KII, DAI Management 32 ANNEXES 33 ANNEX 1: STATEMENT OF WORK SUMMARY Strategy/Project/Activity Name Kenya Niwajibu Wetu (NIWETU) USAID Office USAID/KEA/SPA Implementer(s) DAI Global LLC Contract # AID-OAA-I-13-00013/AID-615-TO-16-00010 Total Estimated Ceiling of the Evaluated Project/Activity (TEC) $20 Million Life of Activity August 2016 – August 2020 Active Geographic Regions Garissa, Isiolo, Mandera, Nairobi, Wajir (five “core counties”) Development Objective(s) (DOs) CDCS 2016–2020 DO1: Devolution Effectively Implemented External or internal evaluation? External I. BACKGROUND A. PROBLEM DESCRIPTION AND CONTEXT Over the past two decades, Kenya has faced a persistent, if not growing, threat of Violent Extremism (VE) from the Somalia-based al-Shabaab (AS) organization and other affiliated Violent Extremist Organizations (VEOs). Kenya’s invasion of neighboring Somalia during the Government of Kenya’s (GoK) Operation Linda Nchi in 2011—in response to repeated AS incursions into Kenya—led to an escalation of AS counterstrikes in Kenya. High-profile attacks against civilian targets proved deadly. The attack on the upscale Westgate Shopping Mall in the nation’s capital left 60 dead. The horrific 2015 attack at Garissa University College resulted in the deaths of 140 students and faculty. Most recently, the Dusit D2 hotel attack in Nairobi in 2019 killed 21 people. AS actions have not been limited to “soft” civilian targets. In 2020, its assault on Camp Simba, a base used by U.S. and Kenyan military forces, in the popular coastal area of Lamu, left three U.S. personnel dead and as many as six aircraft destroyed or damaged. While these brazen and intermittent attacks shocked the public and drew international condemnation, AS and its affiliates perpetrate a more insidious and relentless pattern of small-scale attacks across Kenya’s predominantly rural and arid northeast that leave citizens in a perpetual state of insecurity. In parallel, the GoK took halting steps toward embracing a softer approach to the accelerating VE threat. In September 2016, it launched its National Strategy to Counter Violent Extremism (NSCVE), which proposed nine Countering Violent Extremism (CVE) pillars for the GoK and other CVE stakeholders to focus on for coordinated CVE actions. However, successful, government-led CVE action at the county level proved elusive. There appeared little interest in prioritizing and coordinating CVE activities at the national level, while county and community leaders lacked the skills and resources needed to devise and implement strategies at the grassroots level. At the same time, distrust between security forces and the communities most affected by VE hobbled most local, informal CVE initiatives. In this complicated VE and CVE landscape, NIWETU initiated activities in the last quarter of 2016. 34 B. THE NIWETU INTERVENTION, ITS BASIS, AND THEORY OF CHANGE NIWETU, a four-year activity, was designed to improve the capabilities of communities and government to identify and respond to V) threats in Kenya. The activity focused on five “core counties” which were deemed most vulnerable to VE: Garissa, Isiolo, Mandera, Nairobi, and Wajir counties. NIWETU’s technical approach was grounded in Kenya’s NSCVE, which evolved to pursue a novel “systems-based” methodology that informed the Agency’s 2020 CVE policy and promotes participation of an “inclusive” array of government and non-government stakeholders in CVE efforts. In devising and operationalizing this methodology, NIWETU strengthened capabilities, established linkages between various CVE actors, and encouraged community and government entities to construct their own CVE initiatives. This “all-of￾society” tack is regarded to have proved more effective and more sustainable than past and existing approaches, which had focused narrowly on VE drivers. NIWETU supported Development Objective (DO) 1: Devolution, which was effectively implemented, of USAID/KEA’s Country Development Cooperation Strategy (2016–2020). Specifically, NIWETU targeted Intermediate Result (IR) 1.3: Citizens participate in county affairs; and its sub-IRs 1.3.1: Capacity for civic engagement improved; 1.3.2: Women, youth, and marginalized groups’ participation and representation enhanced; and 1.3.4: Conflict mitigation mechanisms at the county level strengthened. The NIWETU Theory of Change (ToC) stresses the interconnectedness of four result areas that converge to accomplish the program’s objectives: 1) Improved CVE knowledge and skills, 2) Community networks established, 3) Community-led strategies and initiatives, and 4) Increased government capacity. The logical connection holds that: IF a wide variety of local stakeholders—women and youth included—have an enhanced understanding of VE and their leadership skills are strengthened, THEN stakeholders will expand their circle of influence and be able to generate community engagement and support for CVE efforts. And IF CVE stakeholders, organizations, and leaders work together to analyze local VE dynamics and prioritize CVE activities in their contexts and intentionally establish/strengthen functional coordination, THEN community CVE networks can emerge. And IF CVE networks are established and harness unique perspectives and skills of a diverse cross-section of community stakeholders, THEN the community will mobilize to develop and implement initiatives that address local drivers of VE, thereby effectively reducing VE. And IF targeted national and local government officials’ CVE capacity is enhanced, THEN national and county governments will better understand multiple causal layers of VE and effectively collaborate with communities to develop and implement responsive CVE policies and strategies. The cumulative effect of these interrelated interventions serves to accomplish Program Objective 1: Community mobilization to address VE enhanced and Program Objective 2: Government responsiveness to VE improved and, in turn, achieve the program goal: Improve CVE capabilities to identify and respond to VE threats. The below conceptual framework stresses the importance of recognizing the interconnectedness of CVE stakeholders and solutions and ensuring that NIWETU interventions contribute to and strengthen existing networks. Finally, the framework demonstrates the importance of learning in NIWETU’s approach, with research and learning underpinning all NIWETU decision-making. 35 C. NIWETU’S CONCEPTUAL FRAMEWORK D. MONITORING, EVALUATION, AND LEARNING (MEL) PLAN AND OTHER DOCUMENTS NIWETU identified a series of key performance indicators that determined the trajectory and appropriateness of its interventions over the course of the contract as well as measuring and understanding the longer-term magnitude of change. We used twelve custom indicators and one standard indicator to measure project outcomes, along with a custom higher-outcome indicator. We reported on some indicators quarterly and on one indicator annually. We disaggregated indicators by location, type, gender, and age where applicable. Learning was a central pillar in Monitoring and Evaluation (M&E) and we progressively used it to inform program implementation. The contractor will be guided by the following target indicators to evaluate NIWETU’s longitudinal CVE outcomes. 1. Number of people who participated in United States Government (USG)-supported events or activities designed to build mass support for CVE. 2. Number of community-based organizations supported by USG assistance (disaggregated by focus on women, youth, and other) 3. Number of domestic NGOs engaged in monitoring or advocacy work on human rights that had received USG support. 4. Number of community members (non-government officials) trained, mentored, or provided Technical Assistance (TA). 5. Number of women, youth, and marginalized individuals who are beneficiaries of USG-funded interventions. 6. Number of local women who participated in a substantive role or position in CVE supported with USG assistance. 36 7. Number of local youth who participated in a substantive role or position in CVE supported with USG assistance. 8. Number of initiatives directly supported by or adopted by county/national government designed in response to community concerns around VE matters. 9. Percentage of county/national officials trained who indicate they are likely to implement what they learned during training. 10. Number of county/national officials trained, mentored, and/or provided TA as a result of USG assistance. 11. Number of county/national entities that received USG support NIWETU did not measure the impact of its activities through an impact indicator—due to insufficient resources to administer a methodologically valid process, especially if attempting to attribute measured change to the Activity. Instead, the NIWETU team proposed a higher-level outcome indicator that collected perception-based data linked to the Activity’s ToC, namely, perceived changes related to community CVE engagement and government responsiveness. This indicator allowed the team to garner valuable data, linked to the Activity’s objectives, on an annual basis. They carried out quantitative baseline and end-line surveys at the start and end of the project, respectively. To the extent possible, NIWETU used a common set of indicators across all activities and monitoring levels to maximize comparative and cumulative analysis. One example included the likeliness “to implement what they learned during training,” an indicator that is used for both government officials and community members. II. PURPOSE OF THE EVALUATION The purpose of this evaluation is to measure the effectiveness of NIWETU’s interventions in achieving its key objectives: (1) Community mobilization to address “VE enhanced” and (2) Government responsiveness to “VE improved.” In addition, the results of this evaluation will be used to highlight lessons and promote learning of effective development approaches that merit replicating in CVE, a relatively new program area across the Agency and in the Peace and Security sector. Of more immediate concern, and with NIWETU’s mid- and end-line evaluations and final report, we will use the results of this evaluation to inform the design of the Mission’s forthcoming CVE activity. III. QUESTIONS, AUDIENCE, AND INTENDED USE A. Evaluation Questions 1. How effective was the ToC in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved? 2. Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to violent extremism under Objective 1? How can these interventions be improved? 3. To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering violent extremism and advancing its J2SR? What do key counterparts perceive to be the positive and negative aspects of the partnership? 4. How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to violent extremism? How can these interventions be improved? B. Audience and Intended Use The primary audience for the information will be USAID/KEA staff, who will use the findings to inform ongoing interventions designed to mitigate conflict and the design of new programs. They also will share evaluation results with a secondary audience—including USAID/Washington, USAID/West Africa, and other key stakeholders that include GoK ministries, departments, and agencies (in particular, the National 37 Counter Terrorism Center, Ministry of Interior, Coordination of National Government, and select county governments) and development partners such as the European Union, the Danish International Development Agency, and the United Kingdom’s Foreign, Commonwealth & Development Office, which have similar CVE programs in Kenya. The contractor should, therefore, consider the needs of the different audiences and structure the evaluation report and recommendations accordingly. IV. EVALUATION DESIGN, METHODOLOGY, AND ANALYSIS A. Evaluation Design USAID seeks the most robust evaluation design and methodological approach that is appropriate for the scope of the project, resources, and audience. A non-experimental evaluation design should be used. The evaluation will be an external, participatory, and iterative learning exercise. It will use participatory methods with a mix of qualitative and quantitative approaches and draw on input from a range of stakeholders and beneficiaries. Where applicable, data must be disaggregated by gender and age. It is anticipated that the evaluation will include, among other tools, open and semi-structured interviews with key informants, focus group discussions with subgrantees and beneficiaries, and a comprehensive desktop review of documents (both external—research and assessments generated by others, government policies, conflict data, etc., and internal—award documents; MEL plans; quarterly reports, etc.). The contractor will be required to provide an inception report to include more details on the specific approaches/methodologies for achieving the purposes for which the evaluation was commissioned and to finalize such methodology with USAID and, to the extent appropriate, the implementing partner. B. Data Collection Methodology Methodological triangulation is encouraged in this evaluation. At a minimum, the following methods should be considered for use in this regard: Document review of secondary data, Focus Group Discussions (FGDs), and Key Informant Interviews (KIIs). 1. Document Review A desktop review of key documents to be provided by USAID/KEA is required. Key documents to be reviewed must include, but are not limited to: • NIWETU award document including program description • NIWETU baseline data and reports • NIWETU MEL Plan • NIWETU progress reports • USAID/KEA Country Development Cooperation Strategy (2016–2020) and Performance Management Plan (including Performance Indicator Reference Sheets) • Kenya’s National Strategy to Counter Violent Extremism, September 2016 • USAID’s Policy for Countering Violent Extremism Through Development, April 2020 • County Action Plans for NIWETU’s five “core counties.” a. Focus Group Discussions (FGDs) and Key Informant Interviews (KIIs) USAID requests that the evaluator complete the following table as part of its detailed design and evaluation plan. Questions Suggested Data Sources* Suggested Data Collection Methods Data Analysis Methods 1. [Insert Evaluation question] 38 Questions Suggested Data Sources* Suggested Data Collection Methods Data Analysis Methods 2. [Insert Evaluation question] 3. [Insert Evaluation question] b. Data Analysis Methods The contractor must have a clear plan for analyzing and triangulating data from various sources to generate high-quality and credible evidence to answer the evaluation questions. The analysis method must be relevant to the data collection methods and tools proposed. The contractor must thoroughly address potential limitations of the methodologies and ways to mitigate them. This information should be presented in the proposal. All conclusions made by the Evaluation Team (ET) must be supported by clear, verified, and detailed evidence. Anecdotal evidence will not be considered sufficient for discussion and/or drawing conclusions. V. EVALUATION DELIVERABLES AND TIMELINE 1. Inception Report: The contractor must submit a detailed inception report to USAID. This and all subsequent reports must be emailed to the COR in Microsoft Word and PDF formats. The report shall detail questionnaires and interview protocols and should not exceed 15 pages. The inception report should detail the evaluators’ understanding of what is being evaluated and why, showing how each evaluation question will be answered by way of proposed data collection methods, sources of data/information, and data collection and analysis procedures. The report will include a stakeholder’s map, the overall evaluation design and evaluation plan, the methodology including sampling techniques and tools to be used, and a detailed work plan to include a schedule. 2. Presentation of preliminary findings: The team will present the major preliminary findings of the performance evaluation to USAID/KEA in Nairobi through a PowerPoint presentation. The debriefings shall include discussions of findings, initial conclusions, and recommendations. 3. Draft Evaluation Report: The contractor must submit a draft evaluation report incorporating comments given during the presentations. 4. Final Report: Based on the provisions of the USAID evaluation policy, a formal and final evaluation report shall be presented to USAID/KEA. The final report shall incorporate USAID and NIWETU’s comments and suggestions. The report will include an executive summary, table of contents, list of acronyms, evaluation design and methodology, findings, conclusions, and recommendations. The report will be submitted in English. The Final Report must not be more than 40 pages, excluding annexes. 5. Presentation of findings: The team will present the major findings of the performance evaluation to USAID/KEA and relevant Embassy counterparts in Nairobi through a PowerPoint presentation. 6. “Popular Version”: The contractor will prepare a summary of the report, not to exceed 15 pages, which will exclude any procurement-sensitive or sensitive but unclassified information. It will be disseminated among implementing partners, government counterparts, other donors, and other stakeholders, as appropriate. The report must meet standards outlined in the USAID evaluation policy. 39 7. Two Page Summary/Factsheet: The contractor will prepare a fact sheet outlining the major findings and recommendations of the performance evaluation for easy reading and distribution. 8. The contractor must submit the report to the DEC within two days of approval of the final report and provide USAID/KEA with the link. VI. EVALUATION TEAM (ET) COMPOSITION The contractor must provide the most effective team composition based on the proposed methodology. Key personnel must have demonstrated relevant prior experience in Africa; familiarity with USAID’s objectives, approaches, and operations is desirable; and prior evaluation/assessment experience is required. In addition, individual team members should have the technical qualifications identified for their respective positions. The team must have sufficient relevant experience with exposure to conflict mitigation and CVE programming, institutional capacity-building, gender, and youth empowerment. The following key personnel positions are required: Evaluation Team (ET) Leader The Team Leader (TL) is ultimately responsible for the overall management of the ET and the final products. In addition, the TL is responsible for coordinating evaluation activities and ensuring the production and completion of an evaluation report in conformance with the scope of work and timelines. The TL will ensure data integrity, high-quality analysis, and written reports. They also are responsible for quality assurance and timeliness of all deliverables. They are responsible for submitting all deliverables. All team members report to the TL. The evaluator will be required to provide a sample of a relevant evaluation in which the TL was the lead to demonstrate their evaluation, critical analysis, and writing skills. Required Qualifications: • Minimum of a master’s degree in any relevant field of study such as Social or Political Science. • Minimum of six years of experience in carrying out and leading ETs. • Minimum of five years of experience with mixed-method approaches. • Demonstrated skills in data analysis. • Evaluation experience in Sub-Saharan Africa, ideally in East Africa. • Experience working with and/or evaluating civil society organizations in an African context. • Labor Category: M&E or Research Specialist. Senior Investigators The senior investigators, together with the TL, will finalize the evaluation methodology; develop the data collection strategy, instruments, and protocols; direct data collection and compilation; engage in KIIs and FGDs; conduct data analysis; and write the report. It is expected that at least one of the two Senior Investigators will be Kenyan or, at very least, an East African with experience in the field of conflict and CVE. Required Qualifications: • Minimum of a master’s degree in any relevant field of study, or a bachelor’s degree with not less than seven years of experience in lieu of the master’s degree. • Minimum five years of experience in conflict mitigation and/or CVE. • Previous experience in participating in at least five other performance evaluations. • Relevant professional experience in Sub-Saharan Africa, ideally in East Africa. • Fluency in Kiswahili is required for at least one of the Senior Investigators; knowledge of other languages spoken on the Coast is ideal but not required. VII. EVALUATION MANAGEMENT A. Logistical Support 40 USAID/KEA will provide relevant documents and information for this assignment. The USAID/KEA Strategic Planning and Analysis (SPA) Office will manage this contract and coordinate with the Democracy, Governance, and Conflict (DGC) office that managed this award. The contractor will be fully responsible for all logistical and secretarial support, including local and regional travel, and will be required to demonstrate ability to obtain any security and medical clearances required by USAID. USAID may assist, as necessary, with making initial contacts with stakeholders for KIIs and focus groups. B. Scheduling USAID/KEA expects this evaluation to take place beginning on/about January 3, 2022; however, the expectation is that preliminary findings will be available in early March 2022, at the latest, to allow for utilization in design of the Mission’s forthcoming CVE activity, Result Area VII (Preventing and Countering Violent Extremism [CVE]) of the Kenya-Inclusive Governance, Accountability, Performance and Participation (Kenya-IGAPP) program. C. REPORT GUIDELINES • The evaluation report should represent a thoughtful, well-researched, and well-organized effort to objectively evaluate what worked in the project, what did not, and why. • The evaluation report shall address all evaluation questions included in the Statement of Work (SOW). • The evaluation report should include the SOW as an Annex. All modifications to the SOW, whether in technical requirements, evaluation questions, ET composition, methodology, or timeline need to be agreed upon in writing by USAID. • Evaluation methodology shall be explained in detail; and all tools used in conducting the evaluation such as questionnaires, checklists, and discussion guides will be included in an annex in the final report. • Evaluation findings will assess outcomes and impact on males and females, as well as on different age groups, particularly on youth/young people. • Limitations to the evaluation shall be disclosed in the report, with attention to the limitations associated with the evaluation methodology (selection bias, recall bias, unobservable differences between comparator groups, etc.). • Evaluation findings should be presented as analyzed facts, evidence, and data rather than anecdotes, hearsay, or the compilation of people’s opinions. Findings should be specific, concise, and supported by strong quantitative or qualitative evidence. • Sources (including people and organizations contacted) of information need to be properly identified and listed in an annex. • Recommendations need to be supported by a specific set of findings. • Recommendations should be action-oriented, practical, and specific, with defined responsibility for the action. • USAID Evaluation Policy standards must be met by the contractor throughout the contract. 41 ANNEX 2: TOOLS KII INTERVIEW TOOL Source: Targeted and direct beneficiary groups—including youth, women, civil society, the private sector, USAID and IP staff, and government and non-government officials. Parameters: KIIs with key staff and beneficiary populations at the national and county levels. Community representation is critical, and KIIs with vulnerable and targeted group members are essential at the county level. EQ 1: How effective was the ToC in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved? Definitions: “Effective”—sufficiently designed to ensure that aspects of all four objective areas of NIWETU were addressed. Primary Data Sources: KIIs with USAID staff and IP staff (focus on IP management and leadership staff and USAID technical counterparts). Secondary Data Sources: Documents from NIWETU and stakeholders (award documents, MEL Plan, reports, baseline and endline surveys), national CVE plans, and county-level CAP plans. Answering EQ 1 requires assessing the extent to which the ToC articulated the desired changes and necessary interventions that are aligned with GoK’s National Strategy to Counter Violent Extremism and its pillars, as well as the priorities and capacities of beneficiaries, communities, sub-grantees, and government and non-government stakeholders to prevent and counter VE. Interview Prompts: Ask for examples of project strategies that supported the NIWETU design. Which strategies and interventions did or did not work, and why? Did the project target the right stakeholders? Were technical project interventions appropriate for the national and local context in this county? What improvements could be made to the project design and the ToC? EQ 2: Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to VE under DO 1? How can these interventions be improved? Definitions: “Effective”—sufficiently designed to ensure that DO 1 was addressed in NIWETU interventions. Primary Data Sources: KIIs with diverse beneficiaries—including youth, women, and government and non￾government officials. Secondary Data Sources: NIWETU’s reports and documents, MEL data, and sources from national and international organizations working on CVE and conflict. 42 Answering EQ 2 requires measuring the extent to which NIWETU has achieved its results and objectives effectively by gathering credible evidence from multiple data sources and methods on the quantity and quality of interventions, VE prevention initiatives undertaken by communities and sub-grantees, and county governments and any internal and external influencing factors to enhance community capabilities to identify and respond to VE. Interview Prompts: Ask for examples of project interventions that were effective at the county level. Which interventions did or did not work, and why. Were technical project interventions appropriate for the local context in this county? Were interventions inclusive enough to reach the most vulnerable members of the community? How were interventions calibrated to reach youth? Women? Vulnerable group members? EQ 3: To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering violent extremism and advancing its J2SR? What do key counterparts perceive to be the positive and negative aspects of the partnership? Definitions: “Extent”—partnerships that have been scaled and have sustained the LOP as well as in the post-project period—and are being carried out at the county or national levels with leveraged local resources. Primary Data Sources: KIIs with diverse beneficiaries—including youth, women, civil society, and government and non-government officials. Secondary Data Sources: Review of award documents, project reports, County Action Plans, MEL Plan, project output, and outcome indicator data. Answering EQ 3 requires reviewing the systems approach and partnership arrangements that NIWETU has employed at the national and county levels, and the strengths and weaknesses of the relationships toward building capacity to prevent and counter VE. This includes reviewing which aspects of the partnership worked well, which did not, and why. Another aspect of the partnership that the evaluation will investigate is the extent to which resources are leveraged from the community, private sector, CSO, and government to CVE capacity and VE prevention. Interview Prompts: Ask for examples of Private Public Partnerships (PPPs) that are continuing today in some form (county level). Which interventions (partnerships) did or did not work, and why. Were partnerships appropriate for the local context in this county? If not, why not? What could have been done differently? Were interventions inclusive enough to include civil society and reach the most vulnerable members of the community? EQ 4: How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to VE? How can these interventions be improved? 43 Definitions: “Effective”—sufficiently designed to ensure that youth at risk for recruitment and radicalization were engaged or benefited from interventions. Primary Data Sources: KIIs with diverse beneficiaries—including youth, private sector, IP project staff, civil society, and government and non-government officials. Secondary Data Sources: Review of award documents, project reports, County Action Plans, MEL Plan, project output, and outcome indicator data. Answering EQ 4 will include looking into how NIWETU worked with formal and informal structures to engage youth and identify and implement youth-focused activities to address VE push and pull factors, CVE knowledge and skills, and networks to build resilience and prevent risk of radicalization and recruitment to VE. The ET will assess NIWETU’s youth engagement strategies and the quality of engagement by soliciting feedback from youth, government officials, CSOs, and communities on what interventions did and did not work, and why. Interview Prompts: Ask for examples of project interventions that targeted youth recruitment and radicalization risks. Which interventions did or did not work, and why. Were partnerships appropriate for the local context in this county? If not, why not? What could have been done differently? Were interventions inclusive enough to include and reach the most vulnerable youth of the community? FGD INTERVIEW TOOL Source: Targeted and direct beneficiary groups—including youth, women, civil society, the private sector, USAID and IP staff, and government and non-government officials. Parameters: Groups of no fewer than three and no more that five. Some FGDs should be gender￾defined as relevant. Mixed beneficiary groups for specific interventions at the county or national levels may make sense based on triangulation with intervention reporting and impact/outcome data. EQ 2: Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to VE under DO 1? How can these interventions be improved? Definitions: “Effective”—defined as sufficiently designed to ensure that DO 1 was addressed in NIWETU interventions. Primary Data Sources: FGDs with targeted and direct beneficiary groups—including youth, women, and government and non-government officials Secondary Data Sources: NIWETU’s reports and documents, MEL data, and sources from national and international organizations working on CVE and conflict. 44 Answering EQ 2 requires measuring the extent to which NIWETU has achieved its results and objectives effectively by gathering credible evidence from multiple data sources and methods on the quantity and quality of interventions, VE prevention initiatives undertaken by communities and sub-grantees, and county governments; and any internal and external influencing factors to enhance community capabilities to identify and respond to VE. Interview Prompts: Ask for examples of project interventions that were effective at the county level. Which interventions did or did not work, and why? Were technical project interventions appropriate for the local context in this county? Were interventions inclusive enough to reach the most vulnerable members of the community? How were interventions calibrated to reach youth? Women? Vulnerable group members? EQ 3: To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering violent extremism and advancing its J2SR? What do key counterparts perceive to be the positive and negative aspects of the partnership? Definitions: “Extent”—partnerships that have been scaled and have sustained the LOP as well as the post-project period and are being carried out at the county or national levels with leveraged local resources. Primary Data Sources: FGDs with targeted and direct beneficiary groups—including civil society, the private sector, and government, and non-government officials. Secondary Data Sources: Review of award documents, project reports, County Action Plans, MEL Plan, project output, and outcome indicator data. Answering EQ 3 requires reviewing the systems approach and partnership arrangements that NIWETU has employed at the national and county levels, and the strengths and weaknesses of the relationships towards building capacity to prevent and counter VE. This includes reviewing which aspects of the partnership worked well, and which did not, and why. Another aspect of the partnership that the evaluation will investigate is the extent to which resources are leveraged from community, private sector, CSO, and government to CVE capacity and VE prevention. Interview Prompts: Ask for examples of PPP partnerships that are continuing today in some form (County level). Which interventions (partnerships) did or did not work, and why. Were partnerships appropriate for the local context in this county? If not why? What could have been done differently? Were interventions inclusive enough to include civil society and reach the most vulnerable members of the community? EQ 4: How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to VE? How can these interventions be improved? 45 Definitions: “Effective”—defined as sufficiently designed to ensure that youth at risk for recruitment and radicalization were engaged or benefited from interventions. Primary Data Sources: FGDs with targeted and direct beneficiary groups—including youth, women, and government and non-government officials. Secondary Data Sources: Review of award documents, project reports, County Action Plans, MEL Plan, project output, and outcome indicator data. Answering EQ 4 will include looking into how NIWETU worked with formal and informal structures to engage youth and identify and implement youth-focused activities to address VE push and pull factors, CVE knowledge and skills, and networks to build resilience and prevent risk of radicalization and recruitment to VE. The ET will assess NIWETU’s youth engagement strategies and the quality of engagement by soliciting feedback from youth, government officials, CSOs, and communities on what interventions did and did not work, and why. Interview Prompts: Ask for examples of project interventions that targeted youth recruitment and radicalization risks. Which interventions did or did not work, and why? Were partnerships appropriate for the local context in this county? If not, why not? What could have been done differently? Were interventions inclusive enough to include and reach the most vulnerable youth of the community? 46 ANNEX 3: STAKEHOLDERS CONTACTED Role Position Location Contact USAID/KEA Senior Advisor for CVE USAID/KEA John Langlois USAID Chief (former) – DGC Office USAID Stephen Brager USAID Chief (former) – Somalia FO USAID Tyler Beckelman USAID/KEA Outreach and Communications Specialist USAID/KEA Irene Angwenyi C4C Champion Nairobi Hajat Muhidin Sonosi C4C Champion Garissa Mohamed Khalif Nunde C4C Champion Isiola Mohamed Noor Ibrahim C4C Champion Mandera Mohamed Billow Salat C4C Champion Wajir Habiba Shuriye Mgmt – DAI Chief of Party Nairobi John Sampson Mgmt – DAI Deputy Chief of Party Nairobi Ibrahim Somo Mgmt – DAI Program Learning Director Nairobi Catherine Mbindyo Mgmt – DAI Senior Grants/Procurement Officer Nairobi Barbara Wasilwa Mgmt – DAI PO, Research & Govt Support Nairobi Irene Ndun'gu Mgmt – DAI Program Staff Nairobi Ronald Ojwang Mgmt – DAI Program Staff Nairobi Yasmin Farah Mgmt – DAI Grants Staff Nairobi Margaret Ronoh Mgmt – DAI Deputy Director of Grants Nairobi Lilian Lamai Mgmt – DAI M&E Staff Nairobi Rashid Abdiaziz Mgmt – DAI M&E Staff Nairobi Hussein Ahmed Hussein Mgmt – DAI Operations Staff Nairobi Irene Juma Mgmt – DAI Operations Staff Nairobi Anhtony Agalo Okelo Mgmt – DAI Operations Staff Nairobi Lawrence Mwaura Mgmt – DAI Operations Staff Nairobi Carol Muriuki Mgmt – DAI Operations Staff Nairobi Claude Katata Mgmt – DAI Operations Staff Nairobi Samuel Mbugua Mgmt – DAI Operations Staff Nairobi Michael Muithya Mgmt – DAI Procurement Staff Nairobi Kevin Birgen Field Staff – DAI Community Mobilization Officer Nairobi Nelly Kiarie Field Staff – DAI Program Development Officer Nairobi Suli Guhad Field Staff – DAI Community Mobilization Officer Garissa Ismail Abdullahi Field Staff – DAI Program Development Officer Garissa Amina Issa Field Staff – DAI Community Mobilization Officer Isiolo Wario Wako Field Staff – DAI Program Development Officer Isiolo Hassan Bagaja Guyo Field Staff – DAI Community Mobilization Officer Mandera Shukri Adan Field Staff – DAI Program Development Officer Mandera Mohamed Edin Field Staff – DAI Community Mobilization Officer Wajir Hanan Abdisamed Field Staff – DAI Program Development Officer Wajir Abdinoor Omar Ali CSO Grant Rift Valley Institute (RVI) Nairobi Mark Bradbury (Grant signatory) 47 Role Position Location Contact CSO Grant Rift Valley Institute (RVI) Nairobi Pauline Skaper (technical contact) CSO Grant Wasafiri Consulting Kenya Ltd. Nairobi Hamish Wilson/Georgia Gilroy (grant signatory) Subcontract _TS Wasafiri Consulting Kenya Ltd. Nairobi Maia Blume CSO Grant County Governance Watch Nairobi Kevin Osido (grant signatory) CSO Grant County Governance Watch Nairobi Rapudo Hawi (technical contact) PS – Grant Kenya Private Sector Alliance (KEPSA) Nairobi Carole Kariuki (grant signatory) PS – Grant Kenya Private Sector Alliance (KEPSA) Nairobi Victor Ogalo/Davilyne Busuru (technical contacts) CSO Grant Baraza La Walimu Kibra (BAWAKI) Nairobi Suleiman Mohamed Kapsun (grant signatory) CSO Grant Baraza La Walimu Kibra (BAWAKI) Nairobi Musa Mwale (technical contact) CSO Grant Green String Network (GSN) Nairobi Angela Yoder-Maina (grant signatory) CSO Grant Green String Network (GSN) Nairobi Bonface Beti (technical contact) County Gov – Grant Nairobi Regional Commissioner's Office (RC) Nairobi Amos Gikaro Mariba/Wilson Njega (grant signatory) County Gov – Grant Nairobi Regional Commissioner's Office (RC) Nairobi Florah Mworoah (technical contact) Subcontract _TS Swordfish Consulting International, LLC Nairobi Jennifer Farrell (grant signatory) Subcontract _TS Swordfish Consulting International, LLC Nairobi William Farrell (technical contact) CSO Grant Kibra Interfaith for Peace and Development Forum (KIfPDF) Nairobi Baria Kassim (grant signatory) CSO Grant Kibra Interfaith for Peace and Development Forum (KIfPDF) Nairobi Abdul Hussein (technical contact) CSO Grant Tanashati Communications Ltd Nairobi Elijah Girimani (grant signatory) CSO Grant Tanashati Communications Ltd Nairobi Robert Ndwiga (technical contact) Nat. Gov – Grant Ministry of Interior and Coordination Nairobi Amos Gatheca (grant signatory) Nat. Gov – Grant Ministry of Interior and Coordination Nairobi Thomas Sakah (technical contact) Nat. Gov – Grant National Counter Terrorism Centre (NCTC) Nairobi Njenga Miiri (grant signatory) Nat. Gov – Grant National Counter Terrorism Centre (NCTC) Nairobi Njenga Miiri (technical contact) CSO Grant Rural Agency for Community Development and Assistance (RACIDA) Nairobi Mohamoud Duale (grant signatory) CSO Grant Rural Agency for Community Development and Assistance (RACIDA) Nairobi Fred Ogeto (technical contact) CSO Grant Malaika Foundation Nairobi Steve Ouma Akoth (grant signatory) 48 Role Position Location Contact CSO Grant Malaika Foundation Nairobi Japheth Oluoch Ogola (technical contact) CSO Grant Kamukunji Community Peace Network (KACPEN) Nairobi Juma Salim (grant signatory) CSO Grant Kamukunji Community Peace Network (KACPEN) Nairobi Omogi Samwel Omondi (technical contact) Nat. Gov – Grant Kenya School of Government (KSG) Nairobi Dr. Emmanuel Kiangani Nat. Gov – Grant Kenya School of Government (KSG) Nairobi Humphrey Mokaya (technical contact) CSO Grant Perspective Media Ltd Nairobi Isabelle Pierrard (grant signatory) CSO Grant Perspective Media Ltd Nairobi Sasha Choulenina (technical contact) County Gov – Grant Kamukunji Deputy County Commissioner's Office Nairobi Moses Lilan (grant signatory) County Gov – Grant Kamukunji Deputy County Commissioner's Office Nairobi Glory Chemutai (technical contact) CSO Grant Kangemi Kitisuru Youth Development Network (KKYDN) Nairobi Fredrick Oyugi (grant signatory) CSO Grant Kangemi Kitisuru Youth Development Network (KKYDN) Nairobi Francis Wambua (technical contact CSO Grant Supreme Council of Kenyan Muslims Garissa (SUPKEM) Garissa and Nairobi Abdullahi Salat (grant signatory – Garissa), Hassan Ole Naado (grant signatory – Nairobi) CSO Grant Supreme Council of Kenyan Muslims Garissa (SUPKEM) Garissa and Nairobi Aden Abdi Awle (technical contact – Garissa), Shahid Amin Mubari (technical contact – Nairobi) CSO Grant Isiolo Peace Link (IPL) Isiolo Abdia Mohamud (grant Signatory) CSO Grant Isiolo Peace Link (IPL) Isiolo Hussein Mursal (technical contact) CSO Grant Sensitization of Community on Radicalization and Empowerment Solutions (SCORES) Isiolo Abdilatif Abdi (grant signatory) CSO Grant Sensitization of Community on Radicalization and Empowerment Solutions (SCORES) Isiolo Halkano Boru(Technical contact) CSO Grant Isiolo Women of Faith (IWOF) Isiolo Halima Dida (grant signatory) CSO Grant Isiolo Women of Faith (IWOF) Isiolo Flora Nthiga (technical contact) County Gov – Grant Office of the Isiolo County Commissioner (CC) Isiolo John Ondego/Haman Shambi (grant signatory) County Gov – Grant Office of the Isiolo County Commissioner (CC) Isiolo Annabelle Gitonga (technical contact) CSO Grant Garissa Mediation Council (GMC) Garissa Mohammed Osman Ibrahim (grant signatory) CSO Grant Garissa Mediation Council (GMC) Garissa Abdullahi Dagane Dayib (technical contact) CSO Grant Rights Organization for Advocacy and Development (ROAD) Garissa Dahabo Diriye Shalle (grant signatory) 49 Role Position Location Contact CSO Grant Rights Organization for Advocacy and Development (ROAD) Garissa Yusuf Mohamed Haji (technical contact) CSO Grant Reach Out Organization (ROO) Garissa Ibrahim Abubakar (grant signatory) CSO Grant Reach Out Organization (ROO) Garissa Abass Abdullahi Arale (technical contact) CSO Grant Royal Media Services (RMS) Garissa Peninnah Gitau (grant signatory) CSO Grant Royal Media Services (RMS) Garissa Leighton Ingati (technical contact) CSO Grant Dadaab District Peace and Development Group (DDPDG) Garissa Osman Ibrahim (grant signatory) CSO Grant Dadaab District Peace and Development Group (DDPDG) Garissa Omar Abdi Garane (technical contact) CSO Grant Access Initiatives for African Development (AID-Africa) Garissa Ibrahim Mohamed Hussein (grant signatory) CSO Grant Access Initiatives for African Development (AID-Africa) Garissa Ahmed Ali Mohamed (technical signatory) CSO Grant Pastoralist Girls Initiative (PGI) Garissa Fatuma Kinsi (grant signatory) CSO Grant Pastoralist Girls Initiative (PGI) Garissa Barrack Bosire (technical contact) CSO Grant Supreme Council of Kenyan Muslims Garissa (SUPKEM) Garissa and Nairobi Abdullahi Salat (grant signatory – Garissa), Hassan Ole Naado (grant signatory – Nairobi) CSO Grant Supreme Council of Kenyan Muslims Garissa (SUPKEM) Garissa and Nairobi Aden Abdi Awle (technical contact – Garissa), Shahid Amin Mubari (technical contact – Nairobi) CSO Grant Wajir Peace and Development Agency (WPDA) Wajir Abdinasir Saman (grant signatory) CSO Grant Wajir Peace and Development Agency (WPDA) Wajir Shalle Sheikh (technical contact) CSO Grant Raia Development Initiative (RDI) Wajir Mohamud Abdullah (grant signatory) CSO Grant Raia Development Initiative (RDI) Wajir Mohamed Ibrahim (technical contact) CSO Grant Focus On Arid Lands & Integrated Development (FOAID) Wajir Maow Gedi Ismail (grant signatory) CSO Grant Focus On Arid Lands & Integrated Development (FOAID) Wajir Hassan Ali Yussuf (technical contact) CSO Grant Wajir Women for Peace (WWFP) Wajir Halima Abdullahi (grant signatory) CSO Grant Wajir Women for Peace (WWFP) Wajir Fatuma Abdullahi (technical contact) CSO Grant Wajir County Youth Bunge Network (WCYBN) Wajir Muktar Idow Elmi (grant signatory) CSO Grant Wajir County Youth Bunge Network (WCYBN) Wajir Hussein Daud Abdi (technical contact) CSO Grant African Social Development Focus (ASDEF) Wajir Abdishakur A. Mohamed (grant signatory) 50 Role Position Location Contact CSO Grant African Social Development Focus (ASDEF) Wajir Michael Mukavana (technical contact) CSO Grant Delta Africa (DA) Wajr Abdirahman Hassan (grant signatory) CSO Grant Delta Africa (DA) Wajr Ahmed Adan (technical contact) CSO Grant Mandera District Peace Committee (MDPC) Mandera Mr. Yussuf Adan Abdi (grant signatory) CSO Grant Mandera District Peace Committee (MDPC) Mandera Noor Abdow Ahmed (technical contact) CSO Grant Focused Approach Development Concern (FADC) Mandera Abdi Rahoy Kassim (grant signatory) CSO Grant Focused Approach Development Concern (FADC) Mandera Abdi Rahoy Kassim (technical contact) CSO Grant Women Care and Concern (WCC) Mandera Rahma Mohamed Ibrahim (grant signatory) CSO Grant Women Care and Concern (WCC) Mandera Siyad Mohamed Ahmed (technical contact) CSO Grant Nomadic Assistance for Peace and Development (NAPAD) Mandera Abdullahi Hersi (grant signatory) CSO Grant Nomadic Assistance for Peace and Development (NAPAD) Mandera Abdullahi Hersi (technical contact) CSO Grant Greenland Aid Development (GLAD) Mandera Suleiman Ibrahim (grant signatory) CSO Grant Greenland Aid Development (GLAD) Mandera Suleiman Ibrahim (technical contact) CSO Grant Focused Approach Development Concern (FADC) Mandera Abdi Rahoy Kassim (grant signatory) CSO Grant Focused Approach Development Concern (FADC) Mandera Abdirahman Derrow Maalim (technical contact) 51 ANNEX 4: DOCUMENT REVIEW The following are the contractual requirements for the document review: Year 1 – NIWETU Deliverable Documents 1. Year 1 Work Plan 2. Year 2 Work Plan 3. Activity Monitoring and Evaluation Plan (AMEP) 4. Branding and Marking Plan 5. Risk Mitigation Plan 6. Complexity Aware Planning (CAP) Cycle Framework 7. Grants Manual 8. Baseline Assessment 9. Y1 Annual Learning Report 10. Quarterly Research Report: “Recruitment Dynamics in Isiolo” 11. FY 2017 Q1, Q2, and Q3 Quarterly Reports 12. FY 2017 Annual Report 13. Biweekly Reports for September 2016 – September 2017 Year 2 – NIWETU Deliverable Documents 1. Year 3 Work Plan 2. NIWETU Revised AMEP 3. Revised Branding and Marking Plan 4. Y2 Annual Learning Report 5. Quarterly Research Report: “Gender Analysis” 6. Quarterly Research Report: “Dynamics in Garissa Boni Forest” 7. Quarterly Research Report: “Impact of School Closure on Vulnerability in Wajir County” 8. Quarterly Research Report: “Assessment of Improved Skills and Knowledge Activity Outcomes” 9. FY 2018 Q1, Q2, and Q3 Quarterly Reports 10. FY 2018 Annual Report 11. Biweekly Reports for October 2017 – September 2018 Year 3 – NIWETU Deliverable Documents 1. Year 4 Work Plan 2. NIWETU Revised AMEP 3. Revised Risk Mitigation Plan 4. Midline Assessment 5. Y3 Annual Learning Report 52 6. Quarterly Research Report: “The County Action Plan Formation Process: Isiolo, Garissa, and Mombasa” 7. Quarterly Research Report: “Comparative Analysis of Community and Government Responses in the Aftermath of the Westgate and 14 Riverside Attacks” 8. Quarterly Research Report: “Understanding NIWETU’s CVE Approach: A Case Study Analysis of Isiolo Engagement” 9. Quarterly Research Report: “Understanding the Networks of NIWETU’s CVE Champions” 10. FY 2019 Q1, Q2, and Q3 Quarterly Reports 11. FY 2019 Annual Report 12. NIWETU Biweekly Reports for October 2018 – September 2019 Year 4 – NIWETU Deliverable Documents 1. Revised Y4 Work Plan 2. NIWETU Revised AMEP 3. NIWETU Close-down Plan 4. Endline Assessment 5. Y4 Annual Learning Report 6. Quarterly Research Report: “Understanding Key Actions for Successful County Action Plan Implementation” 7. Quarterly Research Report: “Understanding Gender-related Outcomes of NIWETU Grants, and a Review of NIWETU’s Gender Plan” 8. CVE Strategic Communications Evaluation 9. FY 2020 Q1, Q2, and Q3 Quarterly Reports 10. Final Report 11. NIWETU Biweekly Reports for October 2019 – August 2020 Other Documents 1. County CVE Action Plans 2. CDCS Kenya 3. USAID Policy for CVE Through Development Assistance, April 2020 53 ANNEX 5: FIELD QUOTES EQ1: How effective was the ToC in addressing the activity’s objectives? What were the challenges and opportunities of the ToC? How can the ToC be improved? Stakeholder Evidence in Quotes 1. IF a wide variety of local stakeholders—women and youth included—have an enhanced understanding of VE and their leadership skills are strengthened. “The Kumekucha program enhanced my skills in talking to the community. I can engage more in workshops and when interacting in sports and theatre too.” FGD_2_Women “My engagement in the project has given me confidence that I’d be able to talk to people and even interact in the community.” FGD_2_Women “Youth who were part of the project are taking up leadership roles in the community. during this elections period, there are thirty aspirants from the community who want to be MCAs. Our intervention has led to this.” KII_08_KACPEN “… [T]here was nobody who was aware about the violence extremism. Nobody knew about radicalizations. Nobody knew about how to counter it, but since now, we got the skills, and we trained the youth over it, we made the collaborations. This is now the perfect thing that had happened at the time, really changed the security situation of the whole area.” KII_45_DDGPDG 2. THEN stakeholders will expand their circle of influence and be able to generate community engagement and support for CVE efforts. “… [W]e succeeded in the relationship between the community and the security coming together – the community raising those issues and the security responding itself is a milestone.” KII_27_Wajir Peace and Development Agency ‘… [We] have done even integrations between refugees and the community. We have done integrations of the community and the government security agencies, and people are now working hand in hand together, and that has improved the security of the area.” KII_45_DDGPDG 3. And, IF CVE stakeholders, organizations, and leaders work together to analyze local VE dynamics, prioritize CVE activities in their context,s and intentionally establish/strengthen functional coordination, “There was a lot of fear, a lot of extrajudicial killings, a lot of issues affecting the community within Wajir but when the program started there was a lot of campaigns against this issue of CVE. We have done what we call mobilization of mentorship for youth, those who have completed high school. We have done football competitions for those who are unemployed youth in Wajir.” KII_26_RDI “… [W]e succeeded in the relationship between the community and the security coming together – the community raising those issues and the security responding itself is a milestone.” KII_27_WPDA “The partnership between the government and the community at the local government has been enhanced by NIWETU. The NIS, police, chiefs, and security committees are relating better with the community. We are represented at the local councils.” KII_08_KACPEN “NIWETU were directly involved in the Isiolo CAP development, which greatly helped the foundation of CVE work, which is another basis used by other organizations in terms of the design and as a way for them to look at their approaches in delivering CVE.” KII_32–33 “NIWETU gave IPL the opportunity to engage with various structures like Nyumba Kumi and Community Policy, to start aligning peace work and to help bridge the enmity gap between religious leaders and institutions vs. the community.” KII_32–33_IPL 54 4. THEN community CVE networks can emerge. And, IF CVE networks are established and harness unique perspectives and skills of a diverse cross-section of community stakeholders, 5. THEN the community will mobilize to develop and implement initiatives that address local drivers of VE, thereby effectively reducing VE. “What we did was when NIWETU ended and we had daily attacks and our schools were closed, what we did was, we came together as Civil Society Organizations. We did a partnership, and we invited the business community and I think at the time we raised two million shillings that we were able to give to the security actors, the reserves, police reserves and the clan elders at the border level and some of the hardest hit towns. So, that program has been there and it’s active up to now. That’s how we have been trying to survive without programs.” KII_27_WPDA 6. And, IF targeted national and local government officials’ CVE capacity is enhanced, THEN national and county governments will better understand multiple causal layers of VE and effectively collaborate with communities to develop and implement responsive CVE policies and strategies. “… [W]e did, uh, build, uh, the capacity of the county departmental heads because there was acute shortage of skills and knowledge and violent extremism issues in the county—because the county has been saying that security is a preserve for the national government, and they had no role in or a mandate in joining us [to] fight the Al￾Shabaab menace.” KII_27_WPDA Challenges – ToC 1. NIWETU project was not flexible enough to support the continuity of established interventions and partnerships. “Maybe we take those stakeholders broadly and permanently because you know, sometimes you are just fetching people, training them, leaving them like that. So, we might need a broader way of getting people even from the rural areas, educating them, training them on it, and then sending to where they come from…” KII_45_DDPG 2. Many stakeholders strongly felt that NIWETU’s focus was limited to a few sub-counties and could not keep up with the changing tactics and strategies applied by terror groups in Wajir County. They called for future NIWETU projects to expand the scope of reach and focus on cross-border areas. “… [B]ecause we don’t know what can happen, because what we have seen is the shifting of these terror groups now to sub-counties that we thought initially were safe. So, we had some incidences in some sub-counties [where] we thought there were no terror groups and those kinds of things. So, we need also to expand that coverage.” KII_27_WPDA 3. NIWETU’s project duration was short and had limited coverage. “If only NIWETU could be given time to continue, we had enough plans on the ground, that was just to implement the work plan that you have done.” KII_45_DDPGD Field Recommendations 1. “To create ways which get the private sector to be more involved, for example in the forums and the consultative meetings around the CAPs, so that their sentiments are considered and incorporated.” KII_15_DAI-Comm Mobiliser “Strategic communications are not well understood within Govt. and not seen as important. To do CVE communication well, it must take the same approach as campaigns such as AIDS took, to spread it across all segments of society so people realize that this is not an issue of Muslims only or Muslims against Christians.” KII_12-13_Tanashati 55 2. “There should be a focus on social healing [and] training for community-based initiatives and organizations, because VE leaves deep wounds in people which only sink deeper if not addressed.” KII_10-11_GSN EQ 2: Which NIWETU interventions were effective in enhancing community capabilities to identify and respond to VE under DO 1? How can these interventions be improved? Stakeholder Evidence in Quotes Trauma Healing 1. “The trainings enlightened us to what was happening in our community. These things were happening right before our eyes, but we did not see because we did not know what to look for. The trauma in this community is deep, and the things they do is because of that. So, now we can recognize some of these signs.” FGD-01_Kamukunji Youth Men CSO – Grants 2. “The students that PGI targeted got a good grasp of CVE, identification, and radicalization, and were able to identify and discuss these pertinent issues both at school and at home. For example, a parent of one of the students came to school wanting to know why the children were being taught issues of CVE, and this was an indicator that the students were having these conversations at home.” KII_52_PGI 3. “In the development of the sub-county plan of action, we mobilized families from the 14 villages in Majengo. Facilitators were attached to the groups where there was a chairperson and a secretary. They identified the issues affecting their villages and aligned them with the national strategy on VE. Some of the issues were on women, youth, and even ideology.” KII_08_KACPEN C4Cs 4. “NIWETU gave the facilitators/volunteers improved ways to express themselves as C4Cs, which then led to people in the community having the courage to speak out on CVE issues. Before, there was a lot of fear around this issue. They feared being targeted by AS and by the police. They were also traumatized and were not in a good place to network with others until they felt there were people who understood their dilemma.” KII_28_C4C 5. “The C4C was a success. There have been CVE discussions happening to date. The champions having been taking advantage of any opportunity to share CVE knowledge.” KII_34_DAI Field Staff Field Recommendations 1. “In [the] future, the ToC will need to be redesigned. The interventions will move more to preventive [sic]. The PCVE dividend is possible with tangible support, e.g., supporting TIVETS, youth programs and operationalizing the laws.” KII_34_DAI Field Staff 2. “Future projects should focus more on grassroots and cross-border areas since counties in Northern Kenya border Somalia and AS groups exploit the porosity of these borders to recruit and radicalize cross-border communities.” KII_45-CSO Grant 3. “The future design should also change because the knowledge and skills component has been dealt with. It needs to move to preventive [sic]. Youth empowerment programs and provision of scholarships for children from poor backgrounds need to be strengthen[ed].” KII_34_DAI Field Staff 4. “Local actors such as the Ward Development Planning Committees need to be engaged when it comes to matters [of] CVE; otherwise the locals will not open up about their experiences and knowledge due to fear. These actors are more accessible to the community in comparison to the County Engagement Forums.” FGD-06-Isiolo 56 EQ 3: To what extent did NIWETU’s systems approach and partnerships with government (national and county), civil society, and the private sector strengthen Kenya’s commitment and capacity for preventing and countering violent extremism and advancing its J2SR? “The initial approach was of course good because, as a CSO, our first communication was with the national government, the National Counter Terrorism Centre, [to] get them to know, get their policies, involve them in the planning of the project and through participation and implementation of the activities.” KII_63_ROAD One of the challenges with the CEF is that it is still not as inclusive as it could potentially be. For example, the private sector is missing. They are a key player in the country that should participate in it. KII_14_GoK “We were supposed to work with the private sector since they are also casualties in VE, for example, Safaricom. The banking sector too should be engaged. The business community tries to work with the government. We should learn on how they would want to be engaged.” KII_16_RACIDA Through NIWETU, the skills and knowledge impact grew, especially with the security personnel and the community members. The knowledge gaps and constraints were greatly reduced, and other organizations have built on this foundation as well. NIWETU’s direct involvement in the Isiolo CAP development greatly helped the foundation of CVE work, which is another basis used by other organizations for design and evaluating their approaches to delivering CVE. KII_32-33_IPL What do key counterparts perceive to be the positive and negative aspects of the partnership? Positive The partnership between the government and the community at the local government has been enhanced by NIWETU. The NIS, police, chiefs, and security committees are relating better with the community. We are represented at the local councils. KII_8_KACPEN “When women started being involved in CVE, we have seen groups formed. In Isiolo, there is one which is proactive and works with returnees. They have also worked with the security agencies on the same. The same has happened in Kamukunji where the women are more vocal on CVE.” KII_16_RACIDA “KSG and NIWETU has [sic] been one of the key partnerships created. The Security Management Institute was created as a result. It is alive and vibrant. This has been meant to ensure that CVE is mainstreamed in government and civil servants have a point where they can get more CVE knowledge. The institute now has a director, and a different programme (CREATE) is funding it now. This collaboration has also seen the Ministry of Interior and Coordination of Government, NCTC, and KSG even conduct training to the nonstate actors. There have been training modules developed for the different stakeholders like [the] Office of the Deputy Public Prosecutor, Judiciary, and the National Police Service.” KII_Nat gov_NCTC Negative “The relationship between the community and the police is very fragile. One incident could ruin the trust which has been built. We believe we have enhanced the relationship between the community and the police. However, two years ago, the police killed an innocent person, and the community was very angered. The relationship that has been building disconnected. The police and the community spent a few days in running battles where 5 people died and scores were injured. There is need for continuous police and community engagement. They are also traumatized.” KII_8_KACPEN “The CVE toolkit is an intervention that did not work. As much as VE is VE wherever you go, there was lack of contextualization during the development. There was a feeling among different stakeholders that the toolkit did not speak to the Kenyan context. In future, an external organization like HEDAYA who were leading in the development of the toolkit could review the document but not design it. We saw interventions that were led by local organizations having better understanding of the local issues. A good example is what happened in Kamukunji 57 where KACPEN researched on the VE context, understand the context, and came up with an action plan relevant to the community.” KII_6_DAI Mgt “Future partnerships need to understand the context under which they are operating. NIWETU had a partnership with HEDAYA which did not work out. The company was from Dubai and did not understand the local context. There was no consultation with the government and other key stakeholders. It would have been a good document. People from outside cannot come and tell us what to do. We understand VE in our country better than they do. We have used this learning to advise other programmes (CREATE) against taking the same route. We have told them that while CSOs are doing a great job in CVE, they cannot exist in a vacuum. They must be guided by NCTC. Programming on CVE needs to be anchored on the national strategy with NCTC playing an oversight role. Community stakeholders only need to operationalize what has been approved by NCTC.” KII_21_NCTC “There is a knowledge and skills gap created in the county because of the constant transfers of county administrators and security actors (police officials) who have gone through PCVE trainings or who are working on CVE projects. This forces the local organizations to start from scratch, which is time consuming and frustrating.” FGD_6_KSG TOTs “There is need to have more research on the impact of VE on the private sector. They need to be involved more even in their diversity. They need to lead in their own programming as they are the ones affected by VE. The partnership strategies that have been established need to continue with the coordination. Security is expensive and needs resource allocation. The national government needs the private sector and even the CSOs for coordination. We wanted to work with the private sector. There were organizational limitations that affected the engagement. First, the private sector is very broad. KEPSA also lacked capacity for engagement.” KII_07_DAI Mgt_Irene EQ 4: How effectively did NIWETU identify and work with youth (male and female) at risk of radicalization and recruitment to VE? “The design of NIWETU had the youth in mind. There were deliberate efforts to involve them from the very beginning. Apart from the planning that was for them, they were also the main beneficiaries of the project. They may not have been represented at the policy development stage, but they were also the majority running the CSOs which [we] were implementing.” KII_21_GOK “We have seen the youth being more attentive to counsel after the training. Our interaction with the youth has improved because they listen to us more. We have been called when there were issues involving the youth, and [we] mediated. They listen to us. We have seen this especially during this electioneering period when they are more likely to be engaged in violence.” KII_59_Kibra Interfaith “Some youth who engaged with the project got exposure and as result got permanent jobs: one as a DJ in a radio station and another one is not a model in Mombasa after featuring in the videos. We trained a few youth technical work[er]s, such as lighting on sets [and] makeup. They can use these skills and even include [them] in [their] CVs for [the] future. The experience exposed youth to theatre and acting, and in places like Isiolo that do not experience this, it awakened a[n] interest in something new they didn’t know about before. To date, some still call us to keep in touch in case new positions come up for them.” KII_12-13_Tanashati “The project was targeting out-of-school and youth-at-risk individuals so that they can identify the early signs of radicalization; they are able to stay away or to know the strategies used by the groups for recruitment purposes. Ideally, the reach was around 200 youth across the eight sub-counties, but toward the end, 20 young people were selected from all those areas, and then they were able to design a media … a radio talk show that allows these young people to express themselves in terms of where they came from, and where they are right now.” KII_49_DAI Mgt “NIWETU really worked with the youth. The youth champions in the C4C program was [sic] very specific to the youth. It allowed the youth to design their activities and lead community actions in CVE. The young people involved are CVE influencers in their respective communities. There were also other aspects that promoted the involvement of young people. The Majuto play … included King Kaka where the messaging was directed at the young people. 58 Finally, there was a gender plan and a strategic plan to ensure that both men and women were being included in the interventions.” KII_6_DAI Mgt_ How can these interventions be improved? “Future projects should focus more on grassroots and cross-border areas since counties in Northern Kenya border Somalia and AS groups exploit the porosity of these borders to recruit and radicalize cross-border communities.” KII_45_DDPG “The project’s programs should reach the more remote areas of Garissa and Isiolo; there should be stronger cross￾border programs, covering both internal and international borders to address the challenge of the country’s porous borders and the ease with which AS actors are able to operate within these borders. There is still space to address the issue of women and peace in Garissa, especially because VE actors take advantage of gender dynamics as an entry point.” KII_53_County Gov “Sports and theatre are loved in the community. There is need to target interventions involving this. There are many youths with talents, including dancers and even copywriters. There is a mosque in Kamukunji that has been giving young women and girls skills on make-up application and beautification. This also needs to be strengthened.” FGD_02_Women Kamukunji “Women and youth should be targeted and integrated even more during these projects because they make up [much of] the population; and women hold significant sway in the community. Also, youth are not homogeneous, and targeting them should be with that lens of recognizing their differences and reaching them appropriately.” KII_61_BAWAKI