Promoting Equitable Access to Quality Seeds: The Impact of Social and Behaviour Change Interventions among Smallholder Rice Farmers in Butaleja, Uganda October 2025 ENDLINE EVALUATION REPORT | Page 1 of 43 Table of Contents Acronyms and abbreviations……………………………………………………………. 2 Acknowledgement……………………………………………………………………….. 3 1. INTRODUCTION……………………………………………………………………… 5 1.1 Background…………………………………………………………………………… 5 1.2 Purpose of the endline assessment………………………………………………... 6 1.3 Specific objectives……………………………………………………………………. 6 2. LITERATURE REVIEW………………………………………………………………. 7 2.1 Global trends in improved seeds adoption and SBC…………………………….. 7 2.2 The Structure and constraints of seed systems in Uganda……………………....7 2.3 Gender and youth dimensions in seed systems………………………………….. 8 3. METHODOLOGY……………………………………………………………………... 9 3.1 Study design………………………………………………………………………… 9 3.2 Study area…………………………………………………………………………… 9 3.3 Study population and sampling framework………………………………………. 10 4. RESULTS AND DISCUSSION………………………………………………………..12 4.1 Socio demographic characteristics……………………………………………….. 12 4.2 Project relevance and coherence…………………………………………………. 15 4.3 Effectiveness and efficacy…………………………………………………………. 18 4.3.1 Awareness of improved seed varieties among women and youth……….. 20 4.3.3 Main improved seed varieties adopted by rice farmers……………………. 22 4.3.4 Acreage under improved seed varieties……………………………………...23 4.3.5 Changes in farmers farming practices………………………………………. 24 4.3.6 Farmers self-efficacy assessment of their knowledge and skills in farming…………………………………………………………………………………. 25 4.3.7 Recycling and replacement of improved seeds…………………………….. 27 4.4 Project impact and sustainability………………………………………………….. 28 5. CONCLUSION AND RECOMMENDATIONS………………………………………. 33 6.1 Conclusion…………………………………………………………………………… 33 6.2 Recommendations………………………………………………………………….. 33 CGIAR Endline Evaluation Report | Page 2 of 43 Acronyms and abbreviations CBCC: Center for Behavior Change and Communication FGD: Focus Group Discussion IDI: In-Depth Interview IRRI: International Rice Research Institute KII: Key Informant Interview NARO: National Agricultural Research Organization OECD Organization of Economic Co-operation and Development QDS: Quality Declared Seed SBC: Social and Behavior Change STATA: Statistical Software for Data Analysis | Page 3 of 43 Acknowledgement We gratefully acknowledge the technical guidance and financial support provided by the CGIAR Seed Equal Initiative through the International Rice Research Institute (IRRI) and the Alliance of Bioversity International and CIAT. We extend our sincere appreciation to the Center for Behavior Change and Communication (CBCC) for leading the intervention implementation, and to the Butaleja District Agricultural Office and Extension Officers for their unwavering on-ground support throughout the project. Special thanks are due to the rice farmers of Butaleja, whose active participation and insights provided us with data and needed information for this evaluation report, and to the youth champions who amplified the project's impact within the community. Finally, we recognize the dedicated research assistants and consultants whose expertise ensured rigorous data collection, transcription, and analysis. Citation: Bomuhangi, A., & Yila, J. O. (2025). Promoting equitable access to quality seeds: The impact of social and behaviour change interventions among smallholder rice farmers in Butaleja, Uganda: An endline evaluation report. International Rice Research Institute (IRRI). Kampala, Uganda. CGIAR Endline Evaluation Report | Page 4 of 43 Executive summary This endline evaluation assesses the impact of Social and Behaviour Change (SBC) interventions on equitable access to and adoption of quality rice varieties among smallholder farmers in Butaleja District, Uganda. Using a cross-sectional survey design with quasi-experimental methods, the analysis compares 298 treated farmers against 128 control farmers, supplemented by qualitative data from 7 focus group discussions and 13 key informant interviews. The findings highlighted a positive treatment effect across key metrics. Awareness of improved rice varieties reached 95% among treated farmers, compared with the 80% observed in the control group. Ninety-one (91%) percent of the farmers in the treatment group adopted at least one improved variety compared to 42% at baseline and 73% in the control group. This adoption was not only extensive but also intensive, as treated farmers allocated a larger land area to improved varieties (2.3 acres on average) compared to control farmers (1.7 acres). The interventions also induced a shift towards more intensive and knowledge-based production systems, evidenced by a markedly higher rate of inorganic fertilizer use among treated farmers (70%) compared to control groups (25%). The economic prediction is positive, with farmers forecasting a yield increase of 25.86% per acre attributable to the SBC strategies. Furthermore, the project achieved its equity objectives, successfully enhancing the participation of women and youth and fostering local networks for seed access and knowledge sharing, indicating a strong potential for sustaining these net benefits beyond the project's lifecycle. | Page 5 of 43 1. INTRODUCTION 1.1 Background The agriculture sector plays a major role in Uganda's economy, functioning as an engine for poverty alleviation, income generation for smallholder farmers, and the improving national food and nutritional security (World Bank, 2018). The sector employs approximately 70% of the national workforce and contributes about 25% to the country’s Gross Domestic Product (GDP) (UBOS, 2021; World Bank, 2018). Despite its importance, agricultural productivity in Uganda remains suboptimal, characterized by low-input, rain-fed subsistence farming systems (Ogutu & Qaim, 2019). This productivity constraint is particularly acute among smallholder farmers, who account for an estimated 90% of agricultural production but face a myriad of constraints (Bagamba et al., 2023). A well-documented barrier to enhancing productivity and commercializing smallholder agriculture is the persistently low adoption rates of improved agricultural technologies, particularly quality inputs and high-yielding, climate-resilient seed varieties (Fisher & Kandiwa, 2014; Bagamba et al., 2023). This adoption gap is not merely a technical or agronomic challenge; it is deeply intertwined with complex behavioral, social, and structural factors. Issues such as limited access to information, perceived risk, liquidity constraints, and gendered disparities in resource control significantly inhibit farmers' decisions to adopt new technologies (Doss, 2015; Mengie et al., 2022). These barriers disproportionately affect women and youth, who often operate with more limited access to land, credit, extension services, and productive decision-making power, thereby perpetuating cycles of disempowerment and low productivity (Tufa et al., 2022). Recognizing that equitable access to quality seed is a fundamental prerequisite for agricultural transformation, the CGIAR Seed Equal Initiative was conceptualized to achieve this. Work Package 6 of the Seed Equal Initiatives specifically aimed at supporting the equitable delivery of seeds of improved, climate-resilient, market-preferred, and nutritious varieties to last-mile farmers (CGIAR, 2021). Operationalizing this objective, the International Rice Research Institute (IRRI) partnered with the Centre for Behaviour Change and Communication (CBCC) to design and implement strategies for social and behaviour change communication in Eastern Uganda. This project contributed to Outcomes 6.2 and 6.4 of the Seed Equal work package 6, which focused on identifying and scaling strategies for providing women and unreached groups with access to quality seeds. It did so by supporting seed production, information dissemination, and inclusive seed delivery pathways, addressing gender and social constraints as well as the digital divide (CGIAR, 2021). The specific outcomes included enabling women, youth, and disadvantaged groups across 10 sites to access affordable, high-yielding, and resilient varieties; and ensuring that at least five R&D organizations and private seed companies deploy gender-intentional strategies through establishing at least 30 women and youth individuals or collectives in seed production and marketing enterprises. CGIAR Endline Evaluation Report | Page 6 of 43 A formative assessment to establish baseline values for key project indicators was conducted. Informed by these insights, a suite of SBC interventions was designed and implemented to address the identified behavioral, social, and structural constraints to equitable seed access. These interventions leveraged innovative communication strategies, targeted capacity-building, and gender-sensitive approaches for empowering marginalized groups to adopt and utilize quality rice seeds effectively. 1.2 Purpose of the endline assessment The purpose of this assessment was to evaluate the effectiveness and impact of SBC interventions on promoting equitable access to quality seeds among smallholder farmers in Butaleja District. 1.3 Specific objectives The assessment was guided by the following specific objectives, aligned with the standard evaluation criteria of the Organization for Economic Co-operation and Development (OECD, 2019): i. To assess the extent to which the project objectives and design responded to the identified gaps and aligned with project priorities (Relevance) ii. To assess the extent to which the project was compatible with and complemented other interventions in the project sites, including linkages and coordination with other relevant authorities (Coherence) iii. To assess the extent to which the project achieved its objectives and expected results (outputs and outcomes) (Effectiveness). iv. To determine the extent to which the project approaches and activities produced the intended results in an economic and timely manner (Efficiency) v. To determine the most significant changes, document lessons learnt (positive or negative, intended or unintended) that have occurred on the project site because of the project (Impact) vi. To assess the prospects for sustaining the net benefits of the project beyond the current donor funding (sustainability) vii. To evaluate the participation of women and youth in seed production, marketing, and related enterprises viii. To determine shifts in knowledge and social norms and behaviors around improved seed adoption and use ix. To estimate the effect of SBC interventions on intermediate outcomes (Cost, expected crop yield and income) | Page 7 of 43 2. LITERATURE REVIEW 2.1 Global trends in improved seeds adoption and SBC The adoption of improved seed varieties is a key pillar of agricultural transformation and a primary driver of productivity growth, food security, and resilience to climate shocks globally (Singh et al., 2020). These varieties, often bred for higher yield potential, biotic and abiotic stress tolerance, and enhanced nutritional content, represent a low-cost technology with significant potential for improving smallholder livelihoods (Suri, 2011). However, adoption rates are highly heterogeneous, varying considerably across regions, crops, and socio-economic contexts (Acevedo et al., 2020). In Sub-Saharan Africa (SSA), while the adoption rate is increasing, it remains below potential. For major staples like maize, adoption rates were estimated at approximately 60% by 2020, a significant improvement but still lagging behind rates in Asia and the Americas (Cairns et al., 2021). The divergence in adoption patterns is not merely agronomic but is fundamentally explained by a complex web of economic, institutional, and behavioral factors. Barriers are well-documented in the agricultural economics literature and include high perceived risk and uncertainty about performance (Feder et al., 1985), liquidity constraints and lack of access to credit (Croppenstedt et al., 2003), inadequate information and weak extension systems (Anderson & Feder, 2007), and misaligned incentives within seed supply chains (Spielman et al., 2011). Traditional extension models, often focused solely on technology transfer, have proven insufficient to overcome these deeply embedded barriers. This has catalyzed a shift towards Social and Behaviour Change (SBC) strategies, which are designed to address the cognitive, social, and normative drivers of decision-making (Mmbando, 2024; Nchanji et al., 2024). SBC interventions, such as participatory demonstration plots, community-based seed producer models, and innovative use of mass and digital media, function by reducing information asymmetries, facilitating social learning, and building trust, thereby making the adoption of improved seeds a less risky and more socially validated choice (Tewodros et al., 2024). 2.2 The Structure and constraints of seed systems in Uganda Uganda’s agricultural sector is characterized by a dualistic seed system. The formal system, governed by national regulatory frameworks, produces and distributes certified seeds of improved varieties. In parallel, the informal system, comprising farmer-saved seed, local exchange, and grain-for-seed markets, dominates, supplying an estimated 80-90% of all planting material (Akpo et al., 2020). While the informal system offers accessibility and adaptability to local conditions, it often perpetuates the use of genetically degraded and lower-yielding materials, constraining productivity growth (Sperling & McGuire, 2010). CGIAR Endline Evaluation Report | Page 8 of 43 The formal seed system in Uganda faces a myriad of constraints: From a supply-side perspective, key barriers include: limited private sector investment in legume seed production due to perceived low profitability and high transaction costs (Mastenbroek et al., 2021); inadequate infrastructure for seed conditioning and storage; and fragmented distribution networks that fail to reach last-mile farmers, particularly in remote areas (Vandeplas et al., 2022). On the demand side, significant economic and behavioral constraints persist. The high monetary cost of certified seeds relative to grain or saved seeds creates a barrier, especially for resource-constrained farmers. This is compounded by a lack of awareness and persistent misinformation about the returns on investment from improved varieties (Akpo et al., 2020). Furthermore, weak linkages between extension services and farmers limit knowledge diffusion, while cultural preferences for certain varietal traits (e.g., taste, cooking time) can deter adoption even when new varieties are agronomically superior (Ashok et al., 2020). As Bagamba et al. (2023) note, overcoming these challenges requires interventions that simultaneously address market failures and information failures. 2.3 Gender and youth dimensions in seed systems Women are primary custodians of seed in many agricultural communities, and are involved in activities such as seed selection, storage, and preservation (Uduji & Okolo-Obasi, 2023). Despite their contributions, structural barriers limit their access to improved seed varieties and participation in decision-making processes (Westengen et al., 2023). Furthermore, socio-cultural norms often relegate women to subsistence production or lower-value crops. At the same time, men control the income from commercial sales of cash crops, this creates a disincentive for women to adopt improved varieties (Sachs et al., 2020), resulting in a pronounced gender gap in adoption and productivity (Aguilar et al., 2015). Similarly, the young farmers in Uganda are faced with systemic barriers restricting their participation in seed systems (Mulesa et al., 2024). Limited access to land, either through inheritance laws or rental markets, is a primary constraint (Yami et al., 2019). Financial exclusion is also critical, as young farmers often lack collateral to secure loans for inputs like seeds (Anne et al., 2020). Additionally, existing extension services and input delivery models are often not designed to appeal to or effectively reach younger demographics, who may have different information-seeking behaviors and aspirations (Mulesa et al., 2024; White, 2020). | Page 9 of 43 3. METHODOLOGY 3.1 Study design This study employed a quasi-experimental, cross-sectional design to evaluate the impact of Social and Behavior Change Communication (SBCC) interventions on rice farmers in Butaleja District, Uganda. The design incorporated a treatment group, consisting of farmers who directly received the project's SBCC interventions, and a control group, comprising farmers from a similar socio-agroecological context who did not receive the interventions. This approach allowed for the estimation of causal effects by comparing outcomes between the two groups, thereby isolating the impact of the interventions from other confounding factors (Angrist & Pischke, 2008). While a panel data design would have been ideal, the cross-sectional endline survey with a constructed counterfactual provides a robust alternative for impact assessment. 3.2 Study area The research was conducted in Butaleja District, Eastern Uganda (Figure 1). This region was selected as the primary implementation area for the project. Butaleja is characterized by a high concentration of smallholder farmers engaged in the rice value chain. A key constraint in the district is the limited access to high-quality agricultural inputs, which the project aimed to address. The treatment area was Mazimasa Sub-County, and the control area was Himutu Sub-County. These sub- counties were selected for their comparable agro-ecological conditions and farming systems, minimizing external variability. CGIAR Endline Evaluation Report | Page 10 of 43 3.3 Study population and sampling framework The primary study population was smallholder rice farmers. The sampling frame for the treatment group was constructed from verified project beneficiary lists provided by implementing partners and validated by local extension officers. The control group frame consisted of rice-growing households from non-intervention sub-counties, compiled with the assistance of the same extension network to ensure comparability. 3.3.1. Quantitative sample selection A two-stage cluster sampling design was employed for this study. The primary sampling units, Mazimasa (treatment) and Himutu (control) Sub-Counties, were selected purposively based on project implementation status. For the finite population of project-trained farmers in the treatment group (N=940), a minimum sample of 280 was determined using the Yamane (1967) formula for a finite population at a 95% confidence level and a 5% margin of error. For the control group, with an unknown population size, the sample size was calculated using the standard formula for proportions (Cochran, 1977), assuming p=0.5, and yielded 384 households; this was rationally adjusted to 115 due to operational constraints, a recognized practice in field-based research (Lenth, 2001). The initial target of 395 households was subsequently inflated by 10% to account for potential non-response (Kang, 2021), resulting in a planned sample of 426. The final analytical sample, after data cleaning, consisted of 426 respondents (298 treatment, 128 control). 3.3.2. Qualitative sample selection Purposive sampling was used to identify key informants for in-depth interviews (IDIs) and focus group discussions (FGDs) to gain insights into institutional and perceptual changes. Participants were selected from stakeholder groups based on their direct involvement in the project or their position within the rice value chain. The qualitative sample included: 13 in-depth interviews with breeders (NARO), aggregators, district agriculture officers, extension agents, youth champions, community seed producers, and project officers and 7 Focus Group Discussion segregated by gender and location (male, female, and youth groups) in both Mazimasa and Himutu, plus one group of Youth and Women Quality Centers (YWQC) champions. 3.4. Data collection Quantitative data were collected through a structured household survey administered electronically using the Kobo toolbox mobile data capture application. The questionnaire modules captured data on: (a) Socio-demographic characteristics and farm assets, (b) Knowledge, Attitudes, and Practices (KAP) related to improved rice production technologies, (c) Adoption rates of specific technologies promoted by the project (e.g., certified seed, proper spacing, fertilizer use) and (d) Key outcome indicators (e.g., yields, income, input access) while qualitative data was collected using semi-structured interview guides for IDIs and FGDs. These tools explored themes of institutional strengthening, perceived benefits, challenges, sustainability, and gender dynamics. Figure 1: Map of the study area | Page 11 of 43 3.5. Data analysis Quantitative data were analyzed using STATA 17 and Excel. Descriptive statistics (means, proportions, standard deviations) were computed to summarize the characteristics of both treatment and control groups. For qualitative data, Audio recordings from IDIs and FGDs were transcribed verbatim and translated into English. Thematic analysis was conducted using a structured coding framework. Initial codes were developed based on the interview guides and emergent themes from the data. The analysis followed an iterative process of coding, categorization, and theme development to identify consistent patterns, divergent views, and rich narratives explaining the quantitative findings. CGIAR Endline Evaluation Report | Page 12 of 43 4. RESULTS AND DISCUSSION 4.1 Socio-demographic characteristics Table 1 presents the socio-demographic characteristics of rice farmers in Butaleja District, disaggregated by treatment status. Households in the treatment group report a larger average size (9 members) compared to control households (8 members). This may indicate a greater endowment of family labor, a factor associated with enhanced adoption of improved agricultural technologies (Takahashi et al., 2020; Wang et al., 2024). The mean age of household heads is slightly lower in the treatment group (44 years) than in the control group (47 years). Marital status shows minimal variation, with marriage rates nearly identical between treated (88%) and control (89%) household heads. Agricultural self-employment is the dominant occupation across both groups, though its prevalence is higher among treated household heads (98%) relative to the control group (94%). Farmers in the control group reported marginally more farming experience (15 years) than those in the treatment group (14 years). A notable difference is observed in land allocation. Treated farmers allocated an average of 3 acres to improved rice production, compared to 2 acres in the control group. This aligns with established literature suggesting a positive correlation between farm size and the propensity to adopt new technologies (Hu et al., 2022). Qualitative data from Focus Group Discussions (FGDs) suggest that the adoption of improved seeds may also intensify land use. As one participant in Mazimasa noted: “We now plant in lines, which helps us use every inch of our small plots effectively.” Income composition also differed between groups. Treated farmers derived a significantly higher proportion of their household income from rice production (approximately 22%) compared to control farmers (18%). FGDs revealed that this increased income, particularly among women, enhanced financial autonomy and altered intra-household dynamics. Female participants expressed appreciation for their newfound independence, citing an ability to meet personal and children’s needs without spousal reliance. Male participants acknowledged that women's increased contributions alleviated financial pressures and fostered more collaborative decision- making within the household. Finally, the gender distribution of household heads varies between groups. Male- headed households are more prevalent in the control group (76.56%) than in the treatment group (67.45%). | Page 13 of 43 Table 1: Socio-demographic characteristics of rice farmers in Butaleja District, Eastern Uganda Variable Treated (n = 298) Control (n = 128) Pooled (n = 426) Household size (count) 8.6 8.1 8.45 Age of the household head (Years) 44.5 47.6 45.47 Education of the household head None/Illiterate 33.89 12.50 27.46 Adult education or 1 year of education 33.22 12.50 27.00 Religious education 1.01 0.78 0.94 Other education in formal years 31.88 74.22 44.60 Marital status of the household head Single/ Never married 1.68 1.17 Married 88.26 88.73 89.84 Separated 2.68 2.58 2.34 Divorced 1.01 0.94 0.78 Widowed 6.38 6.57 7.03 Polygamous - - - Main occupation of the household head Agriculture self-employed 98.99 94.53 97.65 Agriculture wage labour 0.67 3.13 1.41 Non-agric. self- employment - 0.78 0.23 Non-agric. wage labour - - - Salaried worker - 1.56 0.47 Unemployed 0.34 - 0.23 Sickly/old/not able to work - - - Rice farming experience (Years) 14.16 15.85 14.67 Land under crop production (acres) 2.99 1.80 2.40 Proportion of farm income (%) 28.46 30.98 29.2 Proportion of crop income (%) 21.97 17.77 20.711 Male 67.45 76.56 70.19 Female 32.55 23.44 29.81 CGIAR Endline Evaluation Report | Page 14 of 43 Figure 2 illustrates the formal years of schooling for household heads among rice farmers. The observed difference in educational attainment between the groups is minimal. Both treated and controlled farmers report a mean of 6 years of schooling. Figure 2: Formal years of schooling of household head among rice farmers The age distribution of rice farmers reached in the end-line survey is presented in Figure 3. Adults constituted the significant majority (73%) of respondents. Youth (defined as individuals under 35 years of age) represented nearly a third of the total sample, providing a robust representation of this key project demographic. An apparent disparity exists between treatment groups: the proportion of youth participants was substantially higher in the treated group (32%) compared to the control group (17%). This distribution aligns with the project's target strategy, which prioritized engagement with younger farmers. | Page 15 of 43 Figure 3:Distribution of rice farmers by age group 4.2 Project relevance and coherence Table 2 presents respondent assessments of the relevance and coherence of the project interventions. A large majority of farmers rated the Social and Behavior Change (SBC) strategies as highly relevant, with 65% and 71% of respondents characterizing them as "very effective" in addressing core challenges for rice production, respectively. Disaggregating the data for rice production reveals variation in perceived relevance across demographic groups. Female (69%) and youth (66%) farmers reported higher perceived relevance compared to their male adult counterparts (65%). The promoted technologies and practices demonstrated strong coherence with local farming objectives. Respondents indicated that the improved seed varieties and associated agronomic practices were highly compatible with their household production goals. Qualitative evidence from focus group discussions substantiates these findings. Participants emphasized that the interventions directly addressed critical constraints, including access to improved genetic material and knowledge gaps in sustainable practices. For instance, a youth champion from Mazimasa noted: "We learned about NORO varieties only after CBCC/ENO-SURE began. The project presented opportunities that we were previously unaware of." Furthermore, farmers highlighted specific adaptive benefits, such as the use of drought-resistant seeds and tailored training on improved planting techniques to enhance resilience. Participants from Doho specifically cited the emphasis on timely planting as a crucial strategy for mitigating risks associated with unpredictable weather patterns. Another noted benefit was the improvement in household nutrition, achieved through increased production and access to high-yield, nutrient-rich varieties. CGIAR Endline Evaluation Report | Page 16 of 43 Table 2: Relevance and coherence of interventions Youth (n = 93) Women (n = 67) Men (n=131) Pooled (n=291) Relevance Relevance of SBC strategies in addressing farming challenges Very well 65.59 68.66 64.89 65.98 Moderately well 32.26 31.34 35.11 33.33 Poorly 2.15 - - 0.69 Alignment of promoted seeds and farming practices with households farming goals and needs (1=Yes) 100.00 100.00 97.71 98.97 Coherence Alignment of SBC interventions with other programs Very well 69.89 68.66 64.89 67.35 Moderately well 27.96 31.34 34.35 31.62 Poorly 1.08 - - 0.34 Not at all 1.08 - 0.76 0.69 Conflict/contradictions between the interventions and other programs (1=Yes) 10.75 1.49 8.40 7.56 4.2.1 Reducing gender and youth gaps in the adoption of improved seeds Qualitative findings indicate that the project addressed key gender-based constraints, thereby aiding or facilitating women’s active adoption of improved rice varieties. A female seed entrepreneur in Mazimasa observed, “Women are now able to take care of responsibilities like educating children alongside their husbands, thanks to the project’s support in promoting improved seed varieties’ adoption.” Concurrently, the interventions mitigated youth disengagement from agriculture by creating viable economic opportunities within the sector. As one key informant noted: “The youth used to spend time in non-productive activities like gambling, but now many are involved in farming and related businesses, which has improved their livelihoods” (KII, Mazimasa Sub-County). | Page 17 of 43 4.2.2 Strengthening linkages among actors The project enhanced coordination among actors within the rice value chain. Stakeholders reported improved communication between producers, seed aggregators, and extension services, which reduced transaction costs and information asymmetries. One key informant stated, “Farmers now have direct access to seed sources and technical support, reducing the dependency on unreliable local suppliers” (KII, Mazimasa Cooperative). 4.2.3 Enhancing the SBC interventions Stakeholders expressed strong approval for the Social and Behavior Change (SBC) strategies employed. The use of mass media and participatory group discussions was particularly effective in disseminating agronomic information and encouraging practice adoption. A respondent noted, “The use of radios and group discussions has made it easier for farmers to learn and adopt improved practices.” The deployment of youth champions was also identified as an impactful strategy for peer-to-peer mobilization around improved rice cultivation techniques. To further improve SBC efficacy, respondents recommended reinforcing gender- sensitive programming, particularly by addressing structural barriers such as women’s limited access to land and capital. One seed farmer proposed, “Social and Behavior Change (SBC) interventions should support/propose how women and youth can acquire grants or subsidies to access land and inputs, as this remains a significant challenge” (KII, Seed producer). Additionally, stakeholders advocated for an expansion of localized, on-farm demonstration plots. One district-level official suggested, “More localized demonstrations and improved linkages to reliable seed sources could further boost adoption rates” (KII, Butaleja District). 4.2.4 Intervention’s coherence A significant majority of respondents (76%) reported that the Social and Behavior Change (SBC) interventions complemented existing programs well for rice, with minimal conflict or contradiction with other initiatives. Over 90% of treated respondents perceived no conflict, indicating highly effective project coordination. This coherence was largely driven by strategic alignment with local government policies, particularly those promoting the adoption of Quality-Declared Seeds (QDS), and by integration with public extension services. As one district official noted, “Government policies encourage the use of certified and quality seeds, and this project aligns well by supporting local seed businesses and ensuring that farmers use certified varieties” (KII, Butaleja District). The project demonstrated strong harmonization with other district and national initiatives by occupying a distinct niche. A key informant explained, “CBCC’s behavioral change interventions focused on rice, while other programs targeted other crops, avoiding redundancy and ensuring complementarity” (KII, Butaleja District). The deployment of SBC messaging through existing government extension channels was pivotal in strengthening these synergies. Farmers reported that the project’s training built seamlessly upon prior knowledge, with one noting in an FGD, “The practices we learned here build on what extension officers have been teaching us.” CGIAR Endline Evaluation Report | Page 18 of 43 The regulatory environment for seed certification further enabled the project’s objectives. An extension officer affirmed, “The policy on quality-declared seeds helps farmers access improved varieties, ensuring they achieve better yields” (KII, Extension Officer). Furthermore, the project’s specialized SBC approach minimized functional duplication with the activities of development and research partners, whose efforts were often focused on different technological or crop-specific domains, such as digital agriculture or seed trials for other crops. This clear division of labor ensured that programs complemented rather than competed. 4.3 Effectiveness and efficacy Seventy eight percent (78%) of the respondents rated Social and Behavior Change (SBC) interventions as highly effective in mitigating prevalent farming challenges (Table 3). Project participants anticipated a significant yield increase of 26% per acre attributable to the continued adoption of the promoted agronomic practices. These efficiency gains were projected to reduce production costs by an average of 23% among treated rice farmers. Table 3: Effectiveness and efficacy Criteria Youth (n=93) Women (n = 67) Men (n=131) Pooled (n=291) Addressing farming challenges Very effective 81.72 82.09 74.05 78.35 Somewhat effective 15.05 16.42 25.19 19.93 Neutral 2.15 - 0.76 1.03 Somewhat ineffective 1.08 - - 0.34 Very ineffective - 1.49 - 0.34 Improvement in yield/reduction in farming challenges Yes, significantly 83.87 77.61 76.34 79.04 Yes somewhat 15.05 22.39 23.66 20.62 No change 1.08 - - 0.34 Forecasted yield increase per acre due to SBC interventions (%) 24.93 24.38 27.29 25.86 Forecasted cost reduction per acre due to SBC interventions (%) 22.25 22.50 23.81 23.01 Improving knowledge and skills Very effective 81.72 74.63 72.52 75.95 Somewhat effective 16.13 23.88 25.19 21.99 Not effective 2.15 1.49 2.29 2.06 | Page 19 of 43 Validation from on-farm trials corroborated these perceptions, with demonstration plots recording an average rice yield of 2,321 kg (Figure 4a). Performance varied by variety, with Kafasi 29 achieving the highest mean yield of 2,630 kg, followed by WDR73 and Wita 9 at 2,266 kg and 2,257 kg, respectively. These results indicate the substantial potential for productivity enhancements and cost savings (Figure 4b) through the adoption of improved management techniques and certified seed varieties. Figure 4a: Yield in kgs from actual crop harvests in Butaleja Figure 5b: Costs from actual crop harvests in Butaleja Farmers highlighted during the FGDs that improved seeds yielded significantly better results, with one respondent from Naweyo explaining, "Previously, we were using home-saved seeds with low yields, but now the improved seeds give us much more." CGIAR Endline Evaluation Report | Page 20 of 43 Participants in Mazimasa further emphasized the benefits, stating, "The higher yields mean we can save some rice grains for planting next season and still sell enough to make a profit." FGDs from Doho reported that cost reductions stemmed from more efficient input usage and minimized post-harvest losses. Furthermore, collective improved seed purchase was noted to be reducing cost among the farmers. 4.3.1 Awareness of improved seed varieties among women and youth Awareness of improved seeds was markedly higher among treated women and youth (Figure 5). Treatment group of rice farmers reported about 95% awareness of improved seed varieties relatively higher than their control counterparts who reported a 77% awareness. Similarly, women farmers in the treatment group reported a higher (94%) awareness of the improved seed varieties relative to the women in the control group (79%). Likewise, the youth rice farmers in the treatment group demonstrated more awareness of the improved varieties (nearly 100%) compared to the youth farmers in the control group (72%). Figure 6:Youth and women farmers’ awareness of improved rice seed varieties Figure 6 supports these findings as the results show some notable changes between the baseline and end-line among the rice farmers. For instance, there is an increase in awareness of improved rice varieties among the youth and women farmers from 76% and 72% respectively at baseline to 95% and 94% among the respective group of farmers at end-line. | Page 21 of 43 Figure 6: Improved rice seed varieties’ awareness and adoption status at baseline 4.3.2 Women and youth adoption of improved seed varieties Evaluation data reveal a positive treatment effect on the adoption of improved rice varieties, with particularly pronounced gains among traditionally marginalized groups. As illustrated in Figure 7, adoption rates were markedly higher within treatment groups (91%) compared to controls (73%). For instance, 91% of women rice farmers in the treated group adopted at least one improved variety compared to 79% of women in in the control. This effect was even more substantial among sub-groups; 89% of youth in the treatment group reported adoption, compared to only 73% of youth in the control group, indicating the project's success in mitigating barriers to youth engagement. CGIAR Endline Evaluation Report | Page 22 of 43 Figure 7: Adoption of improved rice seed variety in season 1 of 2023-24 cropping year Qualitative evidence points to a fundamental shift in perceptions, particularly among younger farmers. As one youth participant from Doho noted, “We used to waste time gambling, but now we spend our time in the gardens, and it pays off.” This sentiment was echoed by a youth champion who reframed the economic calculus of agriculture: “Farming is now seen as a business and a way to build our future, not just an activity for survival.” Furthermore, the adoption of improved seeds generated tangible economic benefits that extended beyond yield increases. Female participants in Mazimasa reported enhanced bargaining power in local markets, a direct result of the superior quality and uniformity of their produce. This suggests that technological adoption can improve market positioning and returns for smallholder farmers. The sustained adoption appears to be underpinned by effective knowledge transfer. Focus Group Discussions indicated that project awareness campaigns successfully promoted seed-saving practices, thereby reducing future input costs and securing the continued use of improved varieties. This behavioral change was largely attributed to targeted communication strategies, as a key informant explained: “Radio programs and community workshops transformed farming from a survival activity into a business opportunity.” These interventions effectively reduced information asymmetries and altered the perceived profitability of adopting improved rice varieties. These findings are strongly corroborated by a baseline-to-endline comparison of adoption rates, which reveals substantial and statistically significant increases across all measured demographics. As detailed in Figure 7, the adoption of improved rice varieties among youth farmers surged from a baseline of 43% to 89% at endline. A similarly increase was observed among women farmers, whose adoption rates for improved rice rose from 38% to 91% over the project period. The magnitude of this shift provides empirical evidence for the project's efficacy in overcoming structural and behavioral barriers to technology adoption, particularly for women and youth. 4.3.3 Main improved seed varieties adopted by rice farmers As presented in Figure 8, the adoption of improved rice varieties among treated farmers in Butaleja District was both widespread and concentrated on a few high- performing seeds. The Wita 9 variety demonstrated higher uptake, being adopted by 93% of rice farmers. This was followed by significant adoption rates for other improved varieties: K85 (78%), Supa V-88 (72%), and K8 (50%). The persistence of traditional practices is indicated by the 44% of farmers who continued to plant indigenous varieties, suggesting partial adoption or varietal mixing among a minority of the surveyed population. The clear preference for Wita 9 signals strong farmer acceptance, likely driven by its perceived agronomic performance, marketability, or resilience, and highlights the success of extension efforts in promoting its use. | Page 23 of 43 Figure 8: Main rice varieties grown by smallholder farmers in Butaleja District 4.3.4 Acreage under improved seed varieties The project's interventions induced a significant shift in land allocation decisions, reflecting intensified adoption of improved rice varieties. Farmers in the treatment group allocated a significantly larger average area of 2.3 acres to production using improved seeds, compared to 1.7 acres among control group farmers (Figure 9). This represents not only a 35% increase relative to the control but also a substantial expansion from baseline conditions, where the average land area dedicated to these varieties was between 1.0 and 1.3 acres. The increase in both the proportion of adopters and the scale of adoption, as measured by acreage, provides strong evidence of the intensive margin impact of the SBC strategies. This scaling effect suggests that the interventions successfully altered perceived profitability and reduced risk perceptions, encouraging farmers to commit a greater share of their productive resources to improved rice cultivation. CGIAR Endline Evaluation Report | Page 24 of 43 Figure 9: Average land under improved rice varieties production 4.3.5 Changes in farmers’ farming practices The project induced a shift in input utilization patterns, indicating a transition towards more intensive and knowledge-based production systems. Farmers in the rice treatment group exhibited markedly higher adoption rates of yield-enhancing inputs compared to the control group (Table 4). Specifically, 70.81% of treated rice farmers applied inorganic fertilizer, compared to 25.00% in the control group, while the use of organic amendments (manure) was 40.60% among treated farmers compared to 28.13% among controls. Treated farmers also demonstrated a superior capacity for integrated pest management, with substantially higher application rates of insecticides (52.01%), fungicides (22.82%), and herbicides (13.42%). This pronounced differential in input use highlights the effectiveness of the SBC interventions in overcoming information and technical constraints, leading to more optimal input allocation and enhanced resilience against biotic and abiotic stresses. Table 4: Input used by rice farmers in the 2023-2024 cropping season Input Treatment (n=298) Control (n=128) Pooled (n= 426) Fertilizer 70.81 25.00 57.04 Manure use 40.60 28.13 36.85 Insecticide use 52.01 36.72 47.42 Fungicides use 22.82 21.09 22.30 Herbicide use 13.42 8.59 11.97 | Page 25 of 43 4.3.6 Farmers’ self-efficacy assessment of their knowledge and skills in farming Treatment farmers reported higher levels of self-efficacy across all measured domains of agricultural production, indicating a successful transfer of technical and managerial knowledge. Pronounced disparities emerged between treatment and control groups. For instance, while only 10.16% of control farmers rated their weed and pest control skills as "very good," 35.23% of treatment farmers did so (a more than threefold increase) (Table 5) This pattern of self-assessment held consistently for critical activities such as quality seed selection (40.27% vs. 14.06% "very good"), fertilizer application (32.21% vs. 7.03%), and post-harvest handling (38.26% vs. 14.84%). The most significant competency gaps were observed in oriented skills, including value addition, record keeping, and profit determination, where the proportion of treatment farmers rating their skills as "good" or "very good" was approximately three to four times higher than among controls. This enhanced self- efficacy correlates directly with observed on-farm behavioral changes, such as the adoption of improved drying techniques and basic cost-benefit analysis, suggesting that the capacity-building initiatives effectively built human capital and fostered a more entrepreneurial mindset among participants. The increased self-efficacy highlights the effectiveness of hands-on training and capacity-building initiatives delivered during the project. A farmer from the female rice farmers’ FGD remarked, "We now dry our rice grains on tarpaulins instead of bare ground, which has improved the quality and market value." Moreover, a youth champions in Mazimasa reported that farmers are now tracking input costs and revenues more effectively, enhancing their business acumen. FGDs from Doho noted that increased self-confidence encouraged farmers to share their knowledge with others, fostering a culture of collective learning. Additionally, FGDs revealed that farmers are experimenting with innovative practices, such as organic pest repellents, to further enhance productivity. Table 5: Smallholder farmers’ self-efficacy assessment on their farming knowledge and skills Activity Treatment (n=298) Control (n=128) Pooled (n=426) Land preparation Poor 4.36 24.22 10.33 Highlight row Good 43.29 48.44 44.84 First column Very good 41.95 24.22 36.62 Total row Excellent 10.40 3.13 8.22 Quality seed selection Poor 4.70 39.84 15.26 Good 48.32 45.31 47.42 Very good 40.27 14.06 32.39 Excellent 6.71 0.78 4.93 Planting Poor 3.36 28.91 11.03 Good 45.97 52.34 47.89 Very good 41.28 17.97 34.27 Excellent 9.40 0.78 6.81 Weed and pest Poor 5.03 46.09 17.37 CGIAR Endline Evaluation Report | Page 26 of 43 control Good 54.36 42.97 50.94 Very good 35.23 10.16 27.70 Excellent 5.37 0.78 3.99 Fertilizer application Poor 9.40 61.72 25.12 Good 51.01 31.25 45.07 Very good 32.21 7.03 24.65 Excellent 7.38 - 5.16 Harvest timing Poor 2.35 25.00 9.15 Good 51.68 53.13 52.11 Very good 35.23 19.53 30.52 Excellent 10.74 2.34 8.22 Post-harvest handling and storage Poor 15.77 46.88 25.12 Good 41.28 37.50 40.14 Very good 38.26 14.84 31.22 Excellent 4.70 0.78 3.52 Value addition Poor 22.82 75.78 38.73 Good 39.93 19.53 33.80 Very good 30.54 4.69 22.77 Excellent 6.71 - 4.69 Marketing Poor 25.84 65.63 37.79 Good 39.60 28.13 36.15 Very good 28.19 5.47 21.36 Excellent 6.38 0.78 4.69 Record keeping Poor 29.87 71.88 42.49 Good 34.56 21.09 30.52 Very good 29.87 7.03 23.00 Excellent 5.70 - 3.99 Risk management Poor 32.55 72.66 44.60 Good 35.91 24.22 32.39 Very good 25.50 3.13 18.78 Excellent 6.04 - 4.23 Profit determination Poor 26.17 66.41 38.26 Good 39.60 28.13 36.15 Very good 30.54 5.47 23.00 Excellent 3.69 - 2.58 | Page 27 of 43 4.3.7 Recycling and replacement of improved seeds Analysis of seed recycling and replacement practices reveals nuanced behavioral changes among treatment farmers, indicative of enhanced understanding of seed quality management (Table 6). While an equivalent proportion of farmers in both treatment and control groups (80%) reported recycling improved rice seeds, a critical divergence emerged in replacement behavior. A higher percentage of treated farmers (82%) opted to replace their recycled seeds compared to the control group (78%). More significantly, the duration of recycling exhibited a substantial difference; treatment group farmers recycled improved rice seeds for an average of two seasons, half the average duration of four seasons observed among farmers in the control group. Table 6: Trends in improved rice recycling and replacement Treatment (n=298) Control (n=128) Pooled (n= 426) Recycling of improved seeds 80.54 80.47 80.52 Replacement of recycled seeds 82.08 78.64 81.05 Number of seasons improved seeds recycled 2.17 4.38 2.8 This truncation of the recycling period signifies a shift in risk perception and economic calculation. Farmers in the treatment group appear to have internalized the economic trade-off between the sunk cost of saved seed and the significant yield penalty and genetic degradation associated with prolonged recycling. Qualitative evidence substantiates this; participants cited acquired knowledge of yield reduction and enhanced resilience against biotic stresses as primary motivators for timely replacement "We learned to avoid using the same seeds for many seasons because it reduces the yields. Now we always buy new seeds.". Furthermore, this behavioral shift was facilitated by improved commercial access to genuine seed, suggesting the project’s interventions successfully strengthened last-mile input supply systems and made the procurement of fresh seed a more viable and attractive economic decision for farmers. The major reasons reported by rice farmers behind replacement of recycled improved seeds include reduction in produce physical quality overtime, reduced productivity, transfer of pests and diseases and reduction in crop physiological properties overtime (Figure 10). CGIAR Endline Evaluation Report | Page 28 of 43 Figure 10: Reasons for improved rice seeds replacement 4.4 Project impact and sustainability Over 61% of treated rice farmers rated the project’s impact on household income and food security as highly positive (Table 9), indicating strong perceived economic benefits of the interventions. Quantitatively, the project is forecast to raise crop income by 30% for rice farmers. An analysis of actual harvest data confirms this positive financial impact, revealing an average gross margin of UGX 2,021,667 for rice (Figure 11). | Page 29 of 43 Figure 11: Crop returns from actual crop harvests in Butaleja Of the treated rice farmers, 90.4% reported that the social and behavior change (SBC) interventions positively influenced the decision-making power of women and youth regarding farming practices (Table 7). This enhanced agency, coupled with increased crop yields and income diversification, contributed significantly to improved household resilience. Qualitative evidence substantiates these findings; a female farmer noted, "Now I can sell part of my harvest, pay school fees, and save some money for emergencies," while a participant from Doho stated, "We are no longer worried about bad harvests; the training has made us better prepared." Focus group discussions (FGDs) further emphasized that the increased capacity to invest in farm equipment and children's education was a transformative outcome. An increase in cooperative group activities was also reported, which strengthened community-level economic resilience. CGIAR Endline Evaluation Report | Page 30 of 43 Table 7: Impact and sustainability Criteria Youth (n=93) Women (n = 67) Men (n=131) Pooled (n=291) Impact Overall impact of adopting improved seeds on your household’s income, food security, and well-being? Highly positive 63.44 61.19 59.54 61.17 Positive 29.03 31.34 37.40 33.33 Neutral 1.08 1.49 2.29 1.72 Negative 6.45 5.97 0.76 3.78 Forecasted crop net income due to SBC interventions (%) 31.04 26.19 30.90 29.86 Interventions influence on the decision-making power of women and youth regarding farming practices Yes, positively 87.10 88.06 93.89 90.38 No change 5.38 5.97 5.34 5.50 Negatively 7.53 5.97 0.76 4.12 Sustainability Actions taken since receiving the interventions Adopted improved seeds 94.62 85.07 89.31 90.03 Changed farming practices 77.42 74.63 83.97 79.73 Sought additional information about quality seeds 30.11 28.36 42.75 35.40 Shared information with others 44.09 26.87 30.53 34.02 Number of people shared information with 15.10 19.65 16.08 16.65 Continuous use of the improved seed varieties after the project end 100.00 100.00 99.24 99.66 Developed local sources/ 97.85 94.03 95.42 95.88 | Page 31 of 43 networks ensuring continuous access to the improved seeds/ necessary inputs Have the skills and knowledge acquired during the project enabled you to train or assist other farmers? 100.00 97.01 97.71 98.28 Project sustainability appeared robust. Nearly all treated farmers (99.7%) indicated their intention to continue using improved seed varieties after the project’s conclusion. Furthermore, 95.9% reported having established local networks and sources to ensure continuous access to improved seeds and necessary inputs, a critical factor for maintaining benefits beyond donor funding. The intervention also stimulated significant knowledge diffusion, with participants sharing information with an average of 17 peers. This farmer-to-farmer extension was highlighted by a youth champion from Doho: "Our work doesn’t stop at our gardens. We teach others to do the same, ensuring knowledge spreads." Importantly, 98.3% of respondents felt the skills acquired had enabled them to train or assist other farmers. Strategies proposed by farmers to enhance sustainability include forming cooperative groups to pool resources for bulk seed purchases and strengthening partnerships with local agro- dealers. A key informant interview (KII) with a cooperative leader in Mazimasa confirmed this approach, noting, "Our farmer-to-farmer training ensures the knowledge spreads, even beyond project areas." 4.5: Key impact story Empowering farmers through seed production: Gimbo Nuru’s journey Gimbo Nuru, a 53-year-old rice farmer from Wapala village in Mazimasa Sub-County, Butaleja district, has spent years cultivating rice primarily as a grain crop for sale. However, her perspective shifted dramatically after attending meetings at the Namunasa Youth and Women Quality Centre (YWQC). These gatherings, led by youth champions under the Social and Behavior Change (SBC) interventions, introduced farmers to the concept of rice seed production as a lucrative business venture. Intrigued by the potential, Nuru enrolled in a seed production training organized by the Community-Based Capacity Center (CBCC). The training covered the entire seed production process, including the benefits and profitability of transitioning from grain farming to seed production. For the first time, Nuru began to see farming not just as a means of subsistence but as a pathway to sustainable income generation. Data collected from actual production conditions show that Kafasi 29 rice variety recorded the highest yield of 2630 kg, followed by WDR73 and Wita 9, with yields of 2266 kg and 2257 kg, respectively (Figure 13). CGIAR Endline Evaluation Report | Page 32 of 43 Nuru acknowledges that rice seed production is proving to be far more profitable than cultivating rice for grain. By transitioning to QDS production, she not only stands to increase her income but also contributes to addressing the local challenge of seed unavailability for other farmers. This impact story underscores the transformative potential of targeted agricultural interventions like those delivered through the Social and Behavior Change (SBC) interventions. By empowering farmers like Gimbo Nuru with knowledge, resources, and a market-driven approach, the program is fostering economic resilience, enhancing seed accessibility, and driving sustainable agricultural development in rural communities. | Page 33 of 43 5. CONCLUSION AND RECOMMENDATIONS 6.1 Conclusion The SBC intervention in Butaleja successfully catalyzed equitable, scalable, and sustainable adoption of quality rice seeds. By combining behavioural messaging, peer-led extension, and gender-responsive design, the project overcame key barriers such as information gaps, risk aversion, and social norms that typically limit technology uptake among women and youth. The intervention addressed key barriers to technology adoption, resulting in statistically significant positive treatment effects. The interventions were particularly impactful for traditionally marginalized groups, substantially narrowing gender and youth gaps in awareness, adoption rates, and the scale of adoption (measured by acreage under improved varieties). The project's success is attributed to its high relevance and coherence. The promoted technologies directly addressed local production constraints and were well- aligned with household goals and government policies, such as those promoting Quality-Declared Seeds (QDS). The use of SBC strategies, including mass media, participatory demonstrations, and peer-to-peer learning via youth champions proved effective in disseminating knowledge and shifting perceptions of agriculture from a subsistence activity to a viable business enterprise. The economic impact was substantial. Treated farmers forecast significant yield increases and cost reductions, which were corroborated by actual harvest data showing considerable gross margins. Beyond productivity, the project generated meaningful positive externalities, including enhanced women's decision-making power and intra-household collaboration, improved youth engagement in agriculture, and the development of robust local networks for seed access and knowledge sharing. These factors collectively contributed to greater household resilience and food security. Finally, the project demonstrates a strong potential for sustainability. The intention to continue using improved varieties, coupled with the development of local input networks and a demonstrated capacity for farmer-to-farmer extension, suggests that the benefits are likely to endure beyond the project's lifecycle. 6.2 Recommendations Based on the findings, the following recommendations are proposed to enhance the impact, scalability, and sustainability of similar interventions: i. Scale and deepen SBC interventions with a focus on inclusivity: Future programming should scale these proven SBC strategies to neighboring districts with similar agro-ecological conditions. Interventions must continue to deliberately embed gender and youth-sensitive approaches. To address persistent structural barriers, programs should explore integrating linkages to financial services or matched savings groups to help women and youth overcome capital constraints for accessing land and larger quantities of inputs. CGIAR Endline Evaluation Report | Page 34 of 43 ii. Strengthen last-mile input supply systems: While farmers developed local networks, long-term sustainability requires further formalizing these structures. Project partners should facilitate stronger, more formal linkages between farmer groups/cooperatives and certified seed companies and agro-dealers. Supporting the establishment of rural input stockists or village-based advisory agents, particularly those managed by women and youth, could ensure more reliable and permanent access to quality seeds and other inputs, reducing dependence on project-supported channels. iii. Enhance advanced capacity building: Moving beyond foundational agronomic practices, subsequent training should focus on strengthening farmers' business and managerial acumen. Curricula should be expanded to include modules on post-harvest management, value addition, financial literacy, and detailed record-keeping. This will enable farmers to better capture value from their increased production, optimize profitability, and build resilience against market and climatic shocks. iv. Institutionalize farmer-to-farmer extension: The model of using youth champions and lead farmers was highly effective. Public extension systems and development partners should formally integrate and fund these peer-to- peer models into national agricultural extension policies. Providing certified training and minimal incentives to high-performing champions can create a sustainable, cost-effective system for continuous knowledge dissemination and reinforcement. v. Promote climate-responsive and market-linked varieties: Future efforts should continue to introduce and promote a diverse portfolio of improved varieties, with a specific focus on those offering traits valued by both farmers (drought tolerance, disease resistance) and the market (high milling recovery, grain quality). This dual focus will ensure that productivity gains translate into sustainable income gains, further incentivizing adoption. | Page 35 of 43 Annex 1: Scope of the study Objective Questions To assess the extent to which the project objectives and design responded to the identified gaps and aligned to project priorities (Relevance) QT, QL – KII • Did the project objectives and behaviour change design (models, tools, processes, strategies) respond to the needs of the youth and women small holder farmers? • How important were the behaviour change interventions for youth and women small holder farmers? • How were the gender dimensions considered in the design, implementation, and monitoring? • How did the project enhance opportunities for young people? • How did the project strengthen linkages among various actors? • How well was the project aligned to the policies and priorities of the government? • Give specific examples of how the project addressed the articulated policies and priorities? How would the interventions be designed differently to improve relevance? To assess the extent to which the project was compatible with and complimented other interventions in the project sites including linkages and coordination with other relevant authorities (Coherence) QT, QL – KII • Were the project interventions compatible with and complementary to other interventions in the district, including synergies, interlinkages, harmonization and co-ordination with other relevant initiatives and stakeholders? • How did the project leverage other seed systems initiatives? How did the project approach compliment and synergize interventions by other stakeholders to avoid duplication? To assess the extent to which the project achieved its objectives and expected results (outputs and outcomes) (Effectiveness). QL -KII, FGDs, QT, Project reports • To what extent would you say the project achieved its objectives and expected results: Probe • Improved participation of women to contribute towards inclusive access to new seeds of improved varieties by addressing gender and social barriers to facilitate closing of the gender gaps. • Improved participation of youth to enhance social inclusion and contribute to business opportunities in seed value chains. • Strengthened linkages between value chain actors for information sharing and facilitation of access to quality inputs and collaboration. • Women and youth adopting new seeds of improved varieties. • Women and youth buying high quality seed of improved varieties from different sources. • Average number of seasons that women and youth farmers recycle their seed before replacing it. What were the major factors influencing the achievement or non- achievement of the above results? To determine the extent to which the project approaches and activities produced the intended results in an economic and timely manner (Efficiency) Project reports, QT, KII, FGDs - KII • Did the project approaches and activities produce the intended results in an economic and timely manner? • Do you think the project was good value for money? If not, what alternatives would have delivered the same or better results at lesser cost? • Did the project leverage other resources and capabilities, and if so, to what effect? To what extent were project interventions implemented in the most efficient way compared to other alternatives? To determine the most significant changes, document lessons learnt (positive or Project reports, QT, KIIs, FGDs • What are the most significant lasting changes (positive or negative, intended or unintended) that the project has produced or contributed to and who have been (or are likely to be) impacted and how? CGIAR Endline Evaluation Report | Page 36 of 43 negative, intended or unintended) that has occurred in the project site because of the project (Impact) • Has there been any spillover effect / replication beyond the geographical areas targeted by the project? How did these come about? • Did the project interventions produce any significant unintended consequences, positive or negative, for target participants and groups? What else, if anything, could have been done to strengthen achievement of the project goal and supporting outcomes? To assess the prospects for sustaining the net benefits of the project beyond the current donor funding (sustainability) Project reports, QT, KII, FGDs • What are prospects of sustaining the benefits of the project beyond the current donor funding? • Which strategies were integrated in the project design and implementation to ensure that the project benefits would continue after donor funding has terminated? How well have this sustainability strategies worked? • What are the major factors which could influence the achievement or non-achievement of sustainability? What else, if anything, could have been done to strengthen the sustainability of desired project outcomes? To evaluate the participation of women and youth in seed production, marketing, and related enterprises QT • Effect of SBC interventions on women and youth involvement in producing improved rice/bean seed varieties? • How does participation in different rice and bean value chain nodes enhance women’s role in seed production? • Are there community initiatives that aim to enhance seed access and seed sharing among smallholder farmers? Are there any specific tasks or activities related to seed production that are commonly performed by women and youth To determine shifts in social norms, attitudes, and behaviours around improved seed adoption and use QT, QL-FGD • Are there changes in knowledge, attitudes and practices regarding improved seeds adoption • Changes in social and cultural norms, beliefs, practices, motivations to use seed of choice • Are there persisting barriers that limit women and youth from accessing the necessary resources, e.g., land, water, inputs, tools • Are there changes in the adoption rates of improved seed varieties due to SBC interventions? • Did the campaign influence the target audience’s intent to change their behaviour? • Are individuals showing an increased willingness to adopt the promoted behaviours? • Has the desired behaviour been adopted by the target audience? • Are there observable and measurable changes in behaviour since the campaign started? • Are these changes sustained over time? • Has there been a shift in attitudes or beliefs about the behaviour the campaign promotes? • Are people more aware of the issue or problem addressed by the campaign? • What barriers to behaviour change remain despite the campaign? • Has the campaign successfully addressed specific barriers or amplified enablers of the desired behaviour? • Has the campaign influenced community or societal norms regarding the behaviour? Are people more likely to encourage others to adopt the behaviour? To estimate the effect of SBC interventions on intermediate outcomes (Cost, expected crop yield and income) QT • Are there changes in crop yield due to SBC interventions • Are there changes in crop income due to SBC interventions Are there changes in cost of production due to SBC interventions lesson learnt • What lessons learnt and good practices can be replicated in the future programming and projects. • Which best practices have been adopted and how can they be replicated in future programming or partners for adoption? | Page 37 of 43 • How will beneficiaries and/or other stakeholders utilize the knowledge and skills gained through this programme including complementary activities? • What skills and knowledge should be made accessible to our partners for replication and what negative learning should be avoided? CGIAR Endline Evaluation Report | Page 38 of 43 Annex 2: Endline Survey tools and guides Tool Link Qualitative guide Revised_Eno_Sure_Endline_Assessment_Qualitative_Guide_Revised_ Clean.docx Quantitative tool for Rice Quantitative Tool 1_ Rice Endline Survey-Clean.docx https://centreforbcc.sharepoint.com/:w:/s/Publishablearticles/EToOB9rqqVhAjcC69ZolDh0BpZfyAE-XigF8V6cNJMbG_Q?e=8akEr2 https://centreforbcc.sharepoint.com/:w:/s/Publishablearticles/EToOB9rqqVhAjcC69ZolDh0BpZfyAE-XigF8V6cNJMbG_Q?e=8akEr2 https://centreforbcc.sharepoint.com/:w:/s/Publishablearticles/Ee6KMpKXYG5Gm91Bflawf9sB38q3znPX2Io3MnyWC9rkmg?e=u99Ghk | Page 39 of 43 REFERENCES 1. Acevedo, M., Pixley, K., Zinyengere, N., Meng, S., Tufan, H., Cichy, K., Bizikova, L., Isaacs, K., Ghezzi- Kopel, K., & Porciello, J. (2020). A scoping review of adoption of climate-resilient crops by small-scale producers in low- and middle-income countries. Nature Plants, 6(10), 1231–1241. https://doi.org/10.1038/s41477-020-00783-z 2. Akpo, E., Muricho, G., Lukurugu, G. A., Opie, H., Ojiewo, C. O., & Varshney, R. (2020). Legume seed production for sustainable seed supply and crop productivity: Case of groundnut in Tanzania and Uganda. Journal of Crop Improvement, 34(4), 518–539. https://doi.org/10.1080/15427528.2020.1740368 3. Bagamba, F., Ntakyo, P. R., Otim, G., Spielman, D. J., & Van Campenhout, B. (2023). Policy and performance in Uganda’s seed sector: Opportunities and challenges. Development Policy Review, 41(3), e12665. https://doi.org/10.1111/dpr.12665 4. Cairns, J. E., Chamberlin, J., Rutsaert, P., Voss, R. C., Ndhlela, T., & Magorokosho, C. (2021). Challenges for sustainable maize production of smallholder farmers in sub-Saharan Africa. Journal of Cereal Science, 101, 103274. https://doi.org/10.1016/j.jcs.2021.103274 5. Gilligan, D. O., Kumar, N., McNiven, S., Meenakshi, J. V., & Quisumbing, A. (2020). Bargaining power, decision making, and biofortification: The role of gender in adoption of orange sweet potato in Uganda. Food Policy, 95, 101909. https://doi.org/10.1016/j.foodpol.2020.101909 6. Hertzog, M. A. (2008). Considerations in determining sample size for pilot studies. Research in Nursing & Health, 31(2), 180–191. https://doi.org/10.1002/nur.20247 7. Jayne, T. S., & Sanchez, P. A. (2021). Agricultural productivity must improve in sub-Saharan Africa. Science, 372(6546), 1045–1047. https://doi.org/10.1126/science.abf5413 8. Lenth, R. V. (2001). Some Practical Guidelines for Effective Sample Size Determination. The American Statistician, 55(3), 187–193. https://doi.org/10.1198/000313001317098149 9. Mastenbroek, A., Sirutyte, I., & Sparrow, R. (2021). Information Barriers to Adoption of Agricultural Technologies: Willingness to Pay for Certified Seed of an Open Pollinated Maize Variety in Northern Uganda. Journal of Agricultural Economics, 72(1), 180–201. https://doi.org/10.1111/1477-9552.12395 10. Mulesa, T. H., Otieno, G., Dalle, S. P., Mohamed, M. D., & Westengen, O. T. (2024). Seeding diversity: Enhancing farmers’ access to crop varieties and quality planting materials in Uganda’s seed systems [Report]. Norwegian University of Life Sciences, Ås. https://nmbu.brage.unit.no/nmbu- xmlui/handle/11250/3145800 11. Nchanji, E., Bomuhangi, A., Amuda, A., Mikhala, A., Lengewa, C., Asea, G., Yila, J., L’parnoi, P., Kinyota, P., & Echesa, P. (2024). Social and behavior change and communication strategy for scaling equitable access to rice and common beans quality seeds to smallholder farmers in Butaleja district, Eastern Uganda. Strategy Document. https://hdl.handle.net/10568/159257 12. Rietveld (Anne), A. M., van der Burg (Margreet), M., & Groot (Jeroen), J. C. J. (2020). Bridging youth and gender studies to analyze rural young women and men’s livelihood pathways in Central Uganda. Journal of Rural Studies, 75, 152–163. https://doi.org/10.1016/j.jrurstud.2020.01.020 13. Sachs, C. E., Jensen, L., Castellanos, P., & Sexsmith, K. (2020). Routledge handbook of gender and agriculture. Taylor and Francis. https://doi.org/10.4324/9780429199752 14. Sadikiel Mmbando, G. (2024). The Adoption of Genetically Modified Crops in Africa: The Public’s Current Perception, Regulatory Obstacles, and Ethical Challenges. GM Crops & Food, 15(1), 185–199. https://doi.org/10.1080/21645698.2024.2345401 15. Singh, R. P., Chintagunta, A. D., Agarwal, D. K., Kureel, R. S., & Kumar, S. P. J. (2020). Varietal replacement rate: Prospects and challenges for global food security. Global Food Security, 25, 100324. https://doi.org/10.1016/j.gfs.2019.100324 16. Tewodros, T., Escobar, C. X., Berra, L. S., & Webb Girard, A. (2024). Effectiveness of Elements of Social Behavior Change Activities in Nutrition-Sensitive Agriculture Programs: A Systematic Review. Current Developments in Nutrition, 8(8), 104420. https://doi.org/10.1016/j.cdnut.2024.104420 17. Uduji, J. I., & Okolo-Obasi, E. N. (2023). The impact of CSR on rural women custodians of seed, food and climate change resilience in Nigeria’s Niger Delta region. Journal of Enterprising Communities: People and Places in the Global Economy, 18(3), 441–468. https://doi.org/10.1108/JEC-09-2022-0136 18. Westengen, O. T., Dalle, S. P., & Mulesa, T. H. (2023). Navigating toward resilient and inclusive seed systems. Proceedings of the National Academy of Sciences, 120(14), e2218777120. https://doi.org/10.1073/pnas.2218777120 19. World Bank. (2018). DEVELOPING THE AGRI-FOOD SYSTEM FOR INCLUSIVE ECONOMIC GROWTH. https://documents1.worldbank.org/curated/en/678231542382500879/pdf/132140-WP-P168202- UGANDAECONOMICUPDATE.pdf 20. World Bank Group. (2018). Closing the Potential-Performance Divide in Ugandan Agriculture: Fact Sheet. https://www.worldbank.org/en/country/uganda/publication/closing-the-potential-performance-divide-in- ugandan-agriculture-fact-sheet. 21. Mengie, A., Abdulai, A., & Goetz, R. U. (2022). Impacts of Input Supply, Agricultural Extension and Social Networks on Technology Adoption: Evidence from a Field Experiment in Ethiopia (Doctoral dissertation, Christian-Albrechts Universität zu Kiel). https://doi.org/10.1016/j.foodpol.2020.101909 https://doi.org/10.1126/science.abf5413 https://www.worldbank.org/en/country/uganda/publication/closing-the-potential-performance-divide-in-ugandan-agriculture-fact-sheet https://www.worldbank.org/en/country/uganda/publication/closing-the-potential-performance-divide-in-ugandan-agriculture-fact-sheet CGIAR Endline Evaluation Report | Page 40 of 43 22. Doss, C. R. (2015). Women and agricultural productivity: What does the evidence tell us?. Yale University Economic Growth Center Discussion Paper, (1051). 23. Fisher, M., & Kandiwa, V. (2014). Can agricultural input subsidies reduce the gender gap in modern maize adoption? Evidence from Malawi. Food Policy, 45, 101-111. 24. Ogutu, S. O., & Qaim, M. (2019). Commercialization of the small farm sector and multidimensional poverty. World Development, 114, 281-293. 25. Tufa, A. H., Alene, A. D., Cole, S. M., Manda, J., Feleke, S., Abdoulaye, T., ... & Manyong, V. (2022). Gender differences in technology adoption and agricultural productivity: Evidence from Malawi. World Development, 159, 106027. 26. Uganda Bureau of Statistics (UBOS). (2021). Statistical Abstract. Kampala, Uganda. | Page 41 of 43 CGIAR is a global research partnership for a food-secure future. CGIAR science is dedicated to transforming food, land, and water systems in a climate crisis. Its research is carried out by 13 CGIAR Centers/Alliances in close collaboration with hundreds of partners, including national and regional research institutes, civil society organizations, academia, development organizations and the private sector. www. cgiar.org. We would like to thank the Gender Impact platform for funding this activity. We would like to thank all funders who support this research through their contributions to the CGIAR Trust Fund: www.cgiar.org/funders. To learn more about this Initiative, please visit this webpage. To learn more about this and other Initiatives in the CGIAR Research Portfolio, please visit www.cgiar.org/cgiar-portfolio © 2025 CGIAR System Organization. Some rights reserved. This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 International Licence (CC BYNC 4.0)| | | http://www.cgiar.org/ http://www.cgiar.org/ http://www.cgiar.org/funders http://www.cgiar.org/cgiar CGIAR Endline Evaluation Report | Page 42 of 43