IMPACT OF NATIONAL AGRICULTURAL EXTENSION AND RESEARCH LIAISON SERVICES ADOPTED VILLAGE PROJECT ON BENEFICIARIES’ LIVELIHOOD IN ABIA AND OYO STATES, NIGERIA

CHAPTER ONE: INTRODUCTION

1.1 Background of the Study

Agricultural extension services serve as the critical bridge between agricultural research institutions and farmers, facilitating the transfer of improved technologies, practices, and information from laboratories and experimental stations to rural farm households. In developing countries, where agricultural productivity remains substantially below potential due to knowledge gaps, access to effective extension has been consistently identified as a key determinant of technology adoption, productivity growth, and rural livelihood improvement (Davis et al., 2020). The effectiveness of extension systems, however, varies enormously across countries and contexts, with many public-sector extension systems in sub-Saharan Africa characterized by underfunding, limited reach, weak research-extension linkages, and inadequate attention to the differentiated needs of smallholder farmers (Birner and Resnick, 2019). (Davis et al., 2020; Birner and Resnick, 2019)

Nigeria’s agricultural extension system has undergone multiple reforms and restructuring efforts since independence, reflecting evolving policy priorities and lessons from both domestic experience and international best practices. The system has transitioned from the colonial-era agricultural development approach, through the Training and Visit (TandV) system of the 1980s and 1990s, to the pluralistic, decentralized system that operates today involving federal, state, and local government actors alongside non-governmental organizations and private sector providers (Zakaria and Alhassan, 2018). Despite these reforms, the extension agent-to-farmer ratio in Nigeria remains alarmingly low, estimated at between 1:3,000 and 1:5,000 depending on the region—far below the Food and Agriculture Organization’s recommended ratio of 1:400 for effective service delivery (Federal Ministry of Agriculture and Rural Development, 2020). (Zakaria and Alhassan, 2018; Federal Ministry of Agriculture and Rural Development, 2020)

The National Agricultural Extension and Research Liaison Services (NAERLS), headquartered at Ahmadu Bello University, Zaria, serves as Nigeria’s apex agricultural extension and research liaison institution. Established by decree in 1987, NAERLS has the mandate to coordinate agricultural extension activities nationwide, facilitate linkages between agricultural research institutes and extension delivery systems, develop and disseminate appropriate extension materials, and evaluate extension methodologies and impacts (Olukosi and Isitor, 2019). Over its history, NAERLS has implemented numerous programs and projects aimed at improving agricultural productivity and rural livelihoods, including the Adopted Village Project (AVP) which is the focus of this study. NAERLS operates under the Federal Ministry of Agriculture and Food Security and maintains close relationships with the nation’s network of agricultural research institutes and universities (NAERLS, 2021). (Olukosi and Isitor, 2019; NAERLS, 2021)

The NAERLS Adopted Village Project (AVP) represents a community-based extension approach designed to concentrate extension resources and technical support within selected rural communities, transforming them into demonstration sites for agricultural innovations and best practices. The underlying philosophy of the AVP is that intensive, participatory engagement with a smaller number of communities can produce demonstrable impacts that then serve as catalysts for broader diffusion through neighbor-to-neighbor learning, farmer-to-farmer extension, and observational learning (Adedoyin and Ayinde, 2018). This approach contrasts with the traditional “spraying” extension model that spreads scarce extension resources thinly across large geographical areas, which has been widely criticized as ineffective in generating meaningful changes in farmer knowledge and practices (Akinbile and Ogunlade, 2020). (Adedoyin and Ayinde, 2018; Akinbile and Ogunlade, 2020)

Under the Adopted Village Project, NAERLS selects specific villages based on criteria including accessibility, willingness of community members to participate, agricultural potential, and absence of other intensive extension interventions that would confound impact assessment. Once selected, adopted villages receive a package of extension services that typically includes: regular visits by NAERLS extension agents; training workshops on improved crop, livestock, or fisheries practices; demonstration plots showcasing new crop varieties or agronomic techniques; provision of improved seeds, planting materials, or small livestock; linkages to input suppliers and output markets; and facilitation of farmer groups and cooperatives (NAERLS, 2019). The project emphasizes participatory methods, problem-solving approaches, and building local capacity rather than creating dependency on external support. (NAERLS, 2019)

The conceptual framework underlying the Adopted Village Project draws on several established theories of agricultural innovation diffusion and rural development. The project incorporates elements of diffusion of innovations theory (Rogers, 2003), which emphasizes the role of communication channels, social networks, and opinion leaders in spreading new practices; the farming systems research approach, which recognizes the need to adapt technologies to local conditions and farmer preferences; and participatory extension methodologies that treat farmers as active partners rather than passive recipients of top-down directives (Bamire and Fabiyi, 2019). The project’s emphasis on creating visible demonstration sites and facilitating farmer-to-farmer learning reflects the recognition that social learning and observational learning often drive adoption more effectively than formal instruction alone (Ogunleye and Oladele, 2020). (Bamire and Fabiyi, 2019; Ogunleye and Oladele, 2020)

Livelihoods, as a concept and analytical framework, provides an appropriate lens for evaluating the impact of the Adopted Village Project. The sustainable livelihoods framework, developed by the Department for International Development (DFID) and widely adopted in development research, conceptualizes livelihoods as comprising five capital assets: human capital (skills, knowledge, health), social capital (networks, groups, trust), natural capital (land, water, biodiversity), physical capital (infrastructure, equipment, tools), and financial capital (savings, credit, income flows) (Scoones, 2018). Agricultural extension interventions can affect multiple livelihood capitals simultaneously—for example, by improving farmers’ human capital through training, social capital through group formation, physical capital through technology provision, and financial capital through productivity gains. The AVP’s impact on beneficiaries’ livelihoods must therefore be assessed across multiple dimensions, not solely through agricultural productivity measures (Morse and McNamara, 2019). (Scoones, 2018; Morse and McNamara, 2019)

Abia State, located in Nigeria’s southeastern geopolitical zone, provides one of the two geographical contexts for this study. The state has a predominantly agrarian economy, with farming employing approximately 70% of the working population, primarily at the smallholder level. Major crops include cassava, yam, maize, oil palm, cocoa, and various vegetables, with cassava processing (garri, fufu, starch) representing a significant post-harvest economic activity (Abia State Ministry of Agriculture, 2020). The state falls within the tropical rainforest and derived savannah ecological zones, with abundant rainfall (2,000-2,500mm annually) supporting year-round farming activity. The population is predominantly Igbo, with relatively high population density and land fragmentation characteristic of southeastern Nigeria (Nkwocha and Nwankwo, 2019). The Adopted Village Project was introduced in selected communities in Abia State as part of NAERLS’s efforts to improve agricultural productivity in the region. (Abia State Ministry of Agriculture, 2020; Nkwocha and Nwankwo, 2019)

Oyo State, located in southwestern Nigeria, provides the second study context. The state’s economy is also heavily agricultural, with approximately 65% of the population engaged in farming activities, though the state also has a more diversified economy including significant commerce and services due to the presence of Ibadan, the former regional capital and a major urban center. Major crops in Oyo State include cassava, maize, yam, cowpea, cashew, and various fruits and vegetables, with significant livestock production (poultry, goats, sheep) as well (Oyo State Ministry of Agriculture and Rural Development, 2020). The state falls within the derived savannah and southern Guinea savannah ecological zones, with unimodal rainfall patterns. The population is predominantly Yoruba, with farming systems characterized by a mixture of food crop production and some tree crops (cocoa, cashew) in suitable areas (Fatusin and Ogunleye, 2018). The Adopted Village Project operates in selected rural local government areas of Oyo State, targeting smallholder farmers with extension and technology delivery. (Oyo State Ministry of Agriculture and Rural Development, 2020; Fatusin and Ogunleye, 2018)

Comparing the two states offers analytical value given their contrasting agroecological, socioeconomic, and cultural contexts. Abia State’s rainforest environment supports different cropping systems (e.g., oil palm, cocoyam, plantain) than Oyo State’s savannah environment (e.g., cowpea, cashew, millet), and the farming calendars differ accordingly. Land tenure systems also differ, with more individualized and commercialized land transactions in Oyo State compared to the predominantly lineage-based systems in Abia State (Oladimeji and Adebayo, 2019). In addition, the two states have different extension histories—Abia State was part of the former Imo State and experienced different extension programs, while Oyo State has historically had stronger linkages with IITA (International Institute of Tropical Agriculture) and other research institutions. These differences may influence the baseline conditions against which the AVP’s impact is assessed and the pathways through which impact is achieved (Adebayo and Oladele, 2020). (Oladimeji and Adebayo, 2019; Adebayo and Oladele, 2020)

The theory of change underlying the Adopted Village Project posits that intensive, community-focused extension intervention will produce measurable improvements in beneficiaries’ livelihoods through several causal pathways. First, direct training and demonstration activities increase farmers’ knowledge of improved agricultural practices and their skills in implementing them. Second, provision of improved inputs (seeds, planting materials, livestock) enables farmers to apply new knowledge immediately, overcoming access constraints. Third, facilitation of farmer groups and linkages to input suppliers reduces transaction costs and improves access to quality inputs. Fourth, demonstration plots and visible successes within the community stimulate observational learning and social diffusion of practices beyond direct project beneficiaries (Adediran and Olaniyi, 2019). Fifth, these combined effects should increase agricultural productivity, reduce production risks, enhance market access, and ultimately improve household income, food security, and asset accumulation. However, each causal link in this chain is subject to empirical verification and may be weaker or stronger depending on context (Idowu and Aderinoye-Abdulwahab, 2021). (Adediran and Olaniyi, 2019; Idowu and Aderinoye-Abdulwahab, 2021)

Previous evaluations of agricultural extension projects in Nigeria, including some assessments of NAERLS programs, have produced mixed findings regarding impact. Some studies have reported significant positive effects on technology adoption, productivity, and income, particularly where projects have been implemented with adequate resources, strong farmer participation, and supportive policy environments (Akinola and Adebayo, 2018). Other studies have found limited or no significant impact, attributing failure to factors including inadequate project duration, insufficient follow-up, weak research-extension-farmer linkages, poor infrastructure, and lack of complementary services such as credit and market access (Ogunwale and Ayinde, 2019). Still other evaluations have noted that while projects may produce positive average impacts, benefits are often captured by better-resourced farmers, reinforcing existing inequalities rather than benefiting the poorest and most vulnerable households (Kehinde and Akinbile, 2020). This mixed evidence underscores the importance of context-specific impact assessment that examines not only average impacts but also heterogeneity across different beneficiary groups. (Akinola and Adebayo, 2018; Ogunwale and Ayinde, 2019; Kehinde and Akinbile, 2020)

The Adopted Village Project specifically, as distinct from NAERLS’s broader portfolio of extension activities, has been the subject of limited systematic impact evaluation. Existing studies have been largely descriptive, focusing on project implementation processes and beneficiary satisfaction rather than rigorous quantitative assessment of changes in livelihood outcomes attributable to the project. A study by Ogunfowora and colleagues (2017) examined the AVP in selected northern states and reported positive farmer perceptions but did not employ counterfactual methods to isolate project impact from other confounding factors. Similarly, a NAERLS internal evaluation (2018) documented technology adoption rates among adopted villages but did not include non-adopted comparison villages, making it impossible to attribute observed adoption to the project rather than to other sources of information or innovation (NAERLS, 2018). Thus, a rigorous impact evaluation using appropriate counterfactual methods is long overdue. (Ogunfowora et al., 2017; NAERLS, 2018)

Several methodological challenges complicate the impact assessment of agricultural extension projects like the Adopted Village Project. First, selection bias arises if project villages and beneficiary farmers are not randomly selected but rather chosen based on observable characteristics (e.g., accessibility, farmer motivation) that may also correlate with outcomes. Second, contamination occurs when comparison villages are exposed to project interventions through spillover effects, biasing estimates toward zero. Third, attribution is difficult when multiple interventions (government programs, NGO activities, market changes) occur simultaneously, making it hard to isolate the specific contribution of the AVP. Fourth, heterogeneity of treatment effects means that average impact estimates may mask important variation across different farmer types, cropping systems, or agroecological zones (Gertler et al., 2018). Fifth, short project evaluation timeframes may miss longer-term impacts that only manifest after multiple seasons or years. Addressing these challenges requires careful research design, including appropriate sampling strategies, collection of baseline data (or careful retrospective reconstruction), use of quasi-experimental methods (propensity score matching, difference-in-differences), and analysis of heterogeneity (Khandker et al., 2019). (Gertler et al., 2018; Khandker et al., 2019)

Livelihood impact measures appropriate for evaluating the Adopted Village Project extend beyond simple agricultural productivity indicators to encompass multiple dimensions of household welfare. Key livelihood outcome variables include: farm income (net returns from crop and livestock enterprises); total household income (including off-farm and non-farm sources); asset ownership (productive assets, consumer durables, livestock); food security (months of adequate household food provisioning, dietary diversity); agricultural technology adoption (adoption rates and intensity for improved varieties, practices, and technologies); social capital (membership and participation in groups, trust, collective action); and subjective well-being (self-reported life satisfaction, perceived economic status) (Carletto et al., 2019). A comprehensive impact assessment should examine multiple outcomes because extension projects may affect different aspects of livelihoods unevenly—for example, increasing income without improving food security, or improving technology adoption without affecting asset accumulation (Barrett and Carter, 2020). (Carletto et al., 2019; Barrett and Carter, 2020)

The National Agricultural Extension and Research Liaison Services has operated the Adopted Village Project in various states across Nigeria for over a decade, with the project having been implemented in Abia and Oyo States for several years. Despite this substantial investment of public resources, no rigorous, independent, cross-state impact evaluation has been conducted that employs appropriate counterfactual methods to estimate the project’s causal effects on beneficiary livelihoods. This evaluation gap represents a significant accountability deficit: policymakers, taxpayers, and development partners have limited evidence on whether the project is achieving its intended objectives and whether resources allocated to the AVP represent a good use of scarce public funds relative to alternative extension approaches (Fakoya and Apantaku, 2018). Without rigorous impact evidence, decisions about scaling up, modifying, or discontinuing the project are made on the basis of faith, stakeholder politics, or anecdotal success stories rather than empirical evidence (Omotesho and Ogunlade, 2020). (Fakoya and Apantaku, 2018; Omotesho and Ogunlade, 2020)

Comparing the impact of the Adopted Village Project across Abia and Oyo States offers the opportunity to assess how contextual factors—agroecology, cropping systems, market access, baseline extension coverage, sociocultural conditions—moderate project effectiveness. A project that works well in one context may be less effective in another if its design does not account for local conditions or if complementary conditions (e.g., functioning input markets, accessible output markets) are absent (Moss and Sumberg, 2019). Understanding such cross-contextual heterogeneity in impact can inform more nuanced, context-sensitive project design and targeting decisions. If the AVP is found to be effective in both states, that would support scaling up to other states; if effective only in one, that would suggest the need for adaptation or different implementation models; if ineffective in both, that would call for fundamental redesign or consideration of alternative extension approaches (Takeshima and Liverpool-Tasie, 2019). (Moss and Sumberg, 2019; Takeshima and Liverpool-Tasie, 2019)

In summary, the NAERLS Adopted Village Project represents a significant public investment in community-based agricultural extension in Nigeria, with the explicit goal of improving smallholder farmers’ livelihoods through intensive, participatory technology delivery and capacity building. However, despite the project’s longevity and reach, no rigorous impact evaluation has been conducted to establish its causal effects on beneficiary livelihoods, particularly in Abia and Oyo States where the project has operated for multiple years. This knowledge gap undermines evidence-based policymaking, resource allocation, and program improvement. This study therefore seeks to fill that gap by conducting a rigorous impact assessment of the AVP on beneficiaries’ livelihoods in Abia and Oyo States, using appropriate quasi-experimental methods to estimate causal effects across multiple livelihood dimensions and to examine heterogeneity of impacts across different farmer types and contexts (Oladele and Adetunji, 2021; Omotayo and Adebayo, 2021). (Oladele and Adetunji, 2021; Omotayo and Adebayo, 2021)

1.2 Statement of the Problems

The NAERLS Adopted Village Project has been implemented in various Nigerian states, including Abia and Oyo States, for several years with the explicit objective of improving the livelihoods of smallholder farmers through intensive extension support. Despite the substantial financial and human resources allocated to this project, no rigorous, independent impact evaluation has been conducted to determine whether the project is achieving its intended objectives. Policymakers, project managers, and development partners therefore lack credible evidence on the project’s effectiveness, leaving them unable to make informed decisions about continuation, modification, scaling up, or termination of the program (Ogunfowora et al., 2017; NAERLS, 2018). This evidence gap represents a fundamental accountability failure in the use of public resources for agricultural extension.

A first specific problem is the absence of counterfactual-based impact estimates for the Adopted Village Project in Abia and Oyo States. Most existing assessments of the project have been descriptive, documenting project activities, beneficiary satisfaction, or self-reported technology adoption without comparing outcomes between project beneficiaries and comparable non-beneficiaries. Without a counterfactual—an estimate of what would have happened to beneficiaries in the absence of the project—it is impossible to attribute observed outcomes to the project rather than to other factors such as secular trends, coincident programs, or pre-existing differences between beneficiaries and non-beneficiaries (Gertler et al., 2018). This methodological weakness renders existing assessments largely uninformative for causal inference about project impacts.

A second problem concerns the multidimensional nature of livelihood impacts and the limited evidence on how the Adopted Village Project affects different livelihood outcomes. Even where technology adoption or productivity changes have been documented, the effects on household income, asset accumulation, food security, social capital, and subjective well-being remain largely unexamined. An extension project could increase adoption of improved practices without generating commensurate income gains (if output prices are unfavorable); could increase income without improving food security (if income is controlled by household members who do not allocate it to food); or could benefit some household members (e.g., male household heads) while leaving others (e.g., women, youth) unaffected or even harmed (Barrett and Carter, 2020). A comprehensive livelihood impact assessment examining multiple outcomes is needed.

A third problem concerns the potential heterogeneity of project impacts across different types of beneficiaries. Even if average impacts are positive, the project may not benefit all farmers equally, and could potentially exacerbate existing inequalities if better-resourced farmers (with more land, assets, education, or social connections) capture the lion’s share of benefits. Conversely, the project might be particularly beneficial for resource-poor farmers if it is intentionally targeted and designed to address their specific constraints (Kehinde and Akinbile, 2020). The current evidence base does not permit examination of heterogeneous impacts by farmer characteristics (gender, age, education, farm size, wealth status, etc.), leaving important equity questions unanswered.

A fourth problem concerns the comparison of project impacts across the two distinct agroecological and socioeconomic contexts of Abia and Oyo States. The Adopted Village Project is implemented using a standardized approach across states, but the effectiveness of that standardized approach may vary considerably depending on local conditions—cropping systems, climate, market access, social organization, existing extension infrastructure, and farmer characteristics. It is entirely possible that the project is highly effective in one state but ineffective or even counterproductive in the other. Without cross-state comparative impact estimates, the project cannot be appropriately adapted to local contexts, and decisions about scaling up to other states would be based on averages that may not apply to any particular context (Takeshima and Liverpool-Tasie, 2019).

A fifth problem concerns the mechanisms or pathways through which the Adopted Village Project achieves (or fails to achieve) livelihood impacts. Even if positive impacts are detected, it is important to understand whether they operate through the intended causal pathways—increased knowledge, improved practices, technology adoption, productivity gains, income increases—or through unintended mechanisms. Understanding these pathways is essential for adaptive management: if a pathway is broken (e.g., knowledge increases but technology adoption does not follow due to input access constraints), then the project can be redesigned to address the specific bottleneck. Current evidence does not permit causal pathway analysis for the AVP in Abia and Oyo States (Adediran and Olaniyi, 2019).

A sixth problem concerns the sustainability of project impacts beyond the period of active project implementation. The Adopted Village Project is intended to be time-limited, with intensive support provided for a defined period after which project activities taper off and communities are expected to sustain practices, farmer groups, and diffusion processes independently. However, there is no evidence on whether impacts persist after project withdrawal or whether farmers revert to previous practices, whether farmer groups remain active, or whether social diffusion continues. This sustainability gap is critical for understanding the long-term return on investment in the project (Morse and McNamara, 2019).

A seventh problem concerns the potential displacement effects of the Adopted Village Project. If the project improves outcomes for beneficiary farmers partly by attracting resources—such as extension agent time, improved seeds, or market linkages—away from non-beneficiary communities, then the net impact across the broader population may be smaller than the gross impact on beneficiaries. Alternatively, positive spillovers to neighboring communities through farmer-to-farmer learning could generate net impacts exceeding the direct effects on beneficiaries. Neither displacement nor spillover effects have been estimated for the AVP, so the total social impact of the project remains unknown (Khandker et al., 2019).

An eighth problem concerns the cost-effectiveness of the Adopted Village Project relative to alternative extension approaches. The AVP is resource-intensive, concentrating extension support in a limited number of communities rather than spreading resources thinly across a wider area. This concentration may be justified if the AVP generates sufficiently large impacts to offset its higher per-beneficiary costs compared to conventional extension. Without rigorous impact estimates and cost data, however, it is impossible to compute cost-effectiveness ratios or compare the AVP with alternative extension models such as farmer field schools, agricultural development projects, or private-sector-led extension (Davis et al., 2020). This economic efficiency question is central to resource allocation decisions.

A ninth problem concerns the measurement and definition of the “beneficiary” in project evaluations. The Adopted Village Project works with communities and farmer groups, but within communities, not all farmers participate equally in project activities. Some farmers may attend training, others may not; some may receive inputs, others may not; some may adopt practices, others may not. Evaluations that treat all farmers in adopted villages as “beneficiaries” risk diluting treatment effects and underestimating the project’s impact on actual participants. Conversely, focusing only on active participants may overestimate the project’s population-level impact. The current evidence base does not permit analysis of dose-response relationships—how outcomes vary with intensity of participation (Moss and Sumberg, 2019).

A tenth problem concerns the gender dimension of project impacts. Women play critical roles in Nigerian agriculture, particularly in food crop production, processing, and marketing, but agricultural extension services have historically been biased toward male farmers, with female farmers receiving less extension contact, less access to improved inputs, and fewer opportunities for participation in training. The Adopted Village Project may have exacerbated or reduced these gender disparities, but there is no empirical evidence on gender-differentiated impacts in Abia and Oyo States. Without such evidence, the project cannot be assessed against gender equity objectives, nor can it be redesigned to better serve female farmers (Ogunleye and Oladele, 2020).

An eleventh problem concerns the reliability of self-reported impact data in the absence of baseline measurements. Many project evaluations rely on beneficiary recall of pre-project conditions, which is subject to recall bias (respondents may overstate past difficulties to make the project appear more effective) and rationalization bias (respondents may adjust memories to be consistent with current satisfaction). In Abia and Oyo States, no systematic baseline survey was conducted prior to AVP implementation, so impact estimates must rely either on recall methods (which are problematic) or on quasi-experimental methods using contemporaneous comparison groups. The feasibility and validity of different approaches in this context require careful methodological attention (Carletto et al., 2019).

A twelfth problem concerns the attribution of outcomes to the NAERLS Adopted Village Project specifically, as distinct from other agricultural development interventions operating in the same communities. Both Abia and Oyo States host numerous government programs (e.g., Agricultural Promotion Policy, Fadama projects), NGO interventions, and private sector initiatives that could also affect farmer outcomes. Without careful identification strategies, project impacts may be confounded by these other programs. The current evidence base does not address this attribution challenge, leaving the specific contribution of the AVP unclear (Gertler et al., 2018).

A thirteenth problem concerns the time horizon for impact assessment. Agricultural extension impacts often take multiple growing seasons to materialize, as farmers experiment with new practices, adapt them to local conditions, and decide whether to continue adoption. Cross-sectional evaluations conducted during or immediately after project implementation may miss longer-term impacts and may overstate impacts if adoption is temporary. The AVP has been operating in Abia and Oyo States for several years, providing an opportunity to examine medium-term impacts, but no such assessment has been conducted (Omotesho and Ogunlade, 2020).

In summary, the NAERLS Adopted Village Project in Abia and Oyo States suffers from a profound evaluation deficit. Despite years of implementation and the investment of public resources, there is no rigorous evidence on: whether the project has improved beneficiary livelihoods (causal impact); how impacts vary across different livelihood dimensions, beneficiary types, and state contexts; through what mechanisms impacts are achieved; whether impacts are sustained; whether cost-effectiveness justifies the intensive approach; or how gender and equity dimensions are affected. This evaluation gap constitutes a substantial problem for evidence-based agricultural extension policy and program management in Nigeria, and it is this gap that the present study is designed to address.

1.3 Aim of the Study

The aim of this study is to assess the impact of the National Agricultural Extension and Research Liaison Services (NAERLS) Adopted Village Project on the livelihood outcomes of beneficiary farmers in Abia and Oyo States, Nigeria.

1.4 Objectives of the Study

The specific objectives of this study are to:

1. Describe the socio-economic characteristics of beneficiary and non-beneficiary farm households in the Adopted Village Project areas of Abia and Oyo States.
2. Identify the types of extension services, improved technologies, and capacity-building activities delivered through the Adopted Village Project to beneficiaries.
3. Estimate the impact of the Adopted Village Project on beneficiaries’ agricultural productivity, farm income, and total household income using appropriate counterfactual methods.
4. Assess the impact of the Adopted Village Project on other livelihood dimensions including food security, asset ownership, technology adoption, and social capital.
5. Examine the heterogeneity of project impacts across different beneficiary subgroups (by gender, farm size, wealth status, and state) and identify the factors that constrain or enhance project effectiveness.

1.5 Research Questions

This study seeks to answer the following research questions:

1. What are the socio-economic characteristics of Adopted Village Project beneficiaries and non-beneficiaries in Abia and Oyo States, and how do they differ at baseline?
2. What extension services, technologies, and capacity-building activities have been delivered to beneficiaries through the Adopted Village Project?
3. What is the causal impact of the Adopted Village Project on beneficiaries’ agricultural productivity, farm income, and total household income, after controlling for confounding factors?
4. How does the Adopted Village Project affect other livelihood outcomes including food security, asset ownership, technology adoption, and social capital?
5. Does the impact of the Adopted Village Project vary significantly across beneficiary subgroups (by gender, farm size, wealth status, and state), and what factors explain this variation?

1.6 Research Hypotheses

Hypothesis One

• Null Hypothesis (H₀₁): The Adopted Village Project has no significant impact on the total household income of beneficiary farmers in Abia and Oyo States compared to non-beneficiaries.
• Alternative Hypothesis (H₁₁): The Adopted Village Project has a significant positive impact on the total household income of beneficiary farmers in Abia and Oyo States compared to non-beneficiaries.

Hypothesis Two

• Null Hypothesis (H₀₂): The Adopted Village Project has no significant impact on the agricultural productivity (output per unit area) of beneficiary farmers compared to non-beneficiaries.
• Alternative Hypothesis (H₁₂): The Adopted Village Project has a significant positive impact on the agricultural productivity of beneficiary farmers compared to non-beneficiaries.

Hypothesis Three

• Null Hypothesis (H₀₃): The Adopted Village Project has no significant impact on the food security status (measured by months of adequate food provisioning) of beneficiary households compared to non-beneficiaries.
• Alternative Hypothesis (H₁₃): The Adopted Village Project has a significant positive impact on the food security status of beneficiary households compared to non-beneficiaries.

Hypothesis Four

• Null Hypothesis (H₀₄): The Adopted Village Project has no significant impact on the adoption rate of improved agricultural technologies among beneficiaries compared to non-beneficiaries.
• Alternative Hypothesis (H₁₄): The Adopted Village Project has a significant positive impact on the adoption rate of improved agricultural technologies among beneficiaries compared to non-beneficiaries.

Hypothesis Five

• Null Hypothesis (H₀₅): The impact of the Adopted Village Project does not differ significantly between male and female beneficiaries in Abia and Oyo States.
• Alternative Hypothesis (H₁₅): The impact of the Adopted Village Project differs significantly between male and female beneficiaries in Abia and Oyo States.

1.7 Significance of the Study

This study is significant for multiple stakeholders. First, for NAERLS as the implementing agency, the findings will provide credible evidence on the effectiveness of the Adopted Village Project, identifying strengths to be reinforced and weaknesses to be addressed through program redesign. Second, for the Federal Ministry of Agriculture and Food Security, the study will inform resource allocation decisions—whether to continue, scale up, modify, or discontinue the AVP relative to other extension approaches. Third, for state governments in Abia and Oyo, the findings will guide coordination between federal and state extension programs and identify areas where complementary investments (e.g., infrastructure, market development) could enhance AVP effectiveness. Fourth, for development partners (World Bank, IFAD, FAO, etc.) that have invested in Nigerian agricultural extension, the study will contribute to the evidence base on what works in extension programming in the Nigerian context, informing future project design. Fifth, for the broader agricultural research community, the study will contribute to the literature on extension impact assessment in sub-Saharan Africa, employing rigorous quasi-experimental methods that can be replicated elsewhere. Sixth, for farmer organizations and civil society, the findings will support evidence-based advocacy for more effective, equitable extension services. Seventh, for the farm households themselves, the study’s findings may lead to program improvements that enhance the benefits they receive from extension services. Finally, by examining gender-differentiated impacts, the study will contribute to policy efforts to address gender gaps in agricultural extension access and outcomes.

1.8 Scope of the Study

The geographical scope of this study is limited to selected Adopted Villages in Abia and Oyo States, Nigeria, where NAERLS has implemented the Adopted Village Project for at least three years prior to data collection. In Abia State, the study focuses on villages in Ikwuano, Umuahia North, and Isiala Ngwa South Local Government Areas where the AVP has been operational. In Oyo State, the study focuses on villages in Oyo East, Oyo West, and Atisbo Local Government Areas. The thematic scope focuses specifically on the AVP’s impact on beneficiary livelihoods, encompassing agricultural productivity, income, food security, asset ownership, technology adoption, and social capital. The study does not extend to environmental impacts (e.g., soil health, water use, biodiversity) of changed agricultural practices, nor to impacts on non-farm enterprises except insofar as they are affected by farm-level changes. The respondent scope includes beneficiary farm households in adopted villages, non-beneficiary farm households in adopted villages (to assess within-village spillovers), and non-beneficiary farm households in comparison villages (to establish the counterfactual). The temporal scope covers the period of AVP implementation in each state (ranging from 3 to 7 years depending on when the AVP was introduced), with primary data collected between 2024 and 2025.

1.9 Limitation of the Study

Several limitations inherent in this study should be acknowledged. First, the absence of a randomized controlled trial design means that causal identification relies on quasi-experimental methods (propensity score matching, difference-in-differences where retrospective baseline data are available), which are more vulnerable to selection bias than experimental designs. Second, the lack of prospective baseline data for many beneficiaries means reliance on recall methods (for difference-in-differences with two periods) or cross-sectional comparisons, each with associated biases. Third, contamination of comparison villages through spillover effects from the AVP may attenuate impact estimates toward zero, potentially leading to false conclusions of no impact. Fourth, the study cannot examine very long-term impacts (beyond 7 years) given the timeframe of AVP implementation. Fifth, seasonal variation in agricultural outcomes and food security means that findings are specific to the season of data collection; multiple rounds of data collection across seasons would provide a more complete picture. Sixth, the study focuses only on Abia and Oyo States, so findings may not be generalizable to other states with different conditions. Seventh, measurement error in self-reported outcome variables (income, productivity, food security) may attenuate estimated impacts or reduce precision. Eighth, attrition of beneficiary households from the sampling frame (due to migration, death, or refusal to participate) may bias estimates if attrition is correlated with outcomes. Ninth, the study cannot examine general equilibrium effects—how the project affects input and output prices at the market level—which could affect non-beneficiaries as well. Tenth, political and security conditions in the study areas may affect data collection timing and quality. Despite these limitations, the study will employ best-practice methods for quasi-experimental impact evaluation, conduct robustness checks, and transparently report all assumptions and sensitivity analyses to maximize the credibility of its findings.

1.10 Definition of Terms

Adopted Village Project (AVP): A community-based agricultural extension intervention implemented by the National Agricultural Extension and Research Liaison Services (NAERLS) in selected Nigerian villages. The project provides intensive extension support, technology delivery, training, demonstration plots, and group facilitation to beneficiary communities over a defined period.

Livelihood: The capabilities, assets (including both material and social resources), and activities required for a means of living. In this study, livelihood is operationalized through multiple dimensions including income, asset ownership, food security, technology adoption, and social capital.

Impact: The causal effect of the Adopted Village Project on beneficiary outcomes, defined as the difference between the actual outcome for beneficiaries and the counterfactual outcome (what would have happened to the same beneficiaries in the absence of the project). Impact is distinguished from outputs (direct products of project activities) and outcomes (changes that may be partially attributable to the project).

Beneficiary: A farm household that has participated in Adopted Village Project activities, including receiving extension visits, attending training, obtaining inputs through the project, participating in demonstration plots, or being a member of project-facilitated farmer groups. The study distinguishes between direct beneficiaries (active participants) and indirect beneficiaries (non-participants in adopted villages who may be affected by spillovers).

Counterfactual: The hypothetical outcome for beneficiaries in the absence of the project, against which actual outcomes are compared to estimate impact. In this study, the counterfactual is approximated using outcomes from non-beneficiary households in comparison villages that are statistically similar to beneficiaries on observable characteristics.

Agricultural Productivity: The output of agricultural production per unit of input. In this study, primary measures include crop yield (kg per hectare) for major crops and total factor productivity (where input data permit calculation).

Food Security: A state in which all people at all times have physical, social, and economic access to sufficient, safe, and nutritious food that meets their dietary needs and food preferences for an active and healthy life. This study measures food security through the Household Food Insecurity Access Scale (HFIAS) and months of adequate household food provisioning.

Technology Adoption: The decision by a farm household to use an improved agricultural technology (improved seed variety, fertilizer application method, crop protection practice, soil and water conservation technique, or post-harvest technology) that was promoted through the Adopted Village Project. Adoption is measured as a binary variable (adopted/not adopted) and as intensity (number of technologies adopted).

Social Capital: The networks, norms, and trust that facilitate cooperation and collective action for mutual benefit. In this study, social capital is measured through membership in farmer groups, participation in group activities, perceived trustworthiness of other community members, and engagement in collective action (e.g., group input purchasing, joint marketing).

Difference-in-Differences (DiD): An econometric technique that estimates causal impact by comparing the change in outcomes over time between beneficiaries (treatment group) and non-beneficiaries (comparison group), under the assumption that in the absence of the treatment, both groups would have followed parallel trends.

Propensity Score Matching (PSM): A statistical technique that estimates the probability (propensity score) of a household being a beneficiary based on observable characteristics, then matches each beneficiary with one or more non-beneficiaries with similar propensity scores, creating a balanced comparison group.

Extension Agent: A professional employed by NAERLS, state agricultural development programs, or local government agricultural departments who provides information, training, and technical assistance to farmers. In the AVP context, extension agents conduct regular visits to adopted villages.

Spillover Effect: The indirect impact of the Adopted Village Project on non-beneficiary households or communities, occurring through mechanisms such as farmer-to-farmer learning, demonstration plot observation, or shared market linkages. Positive spillovers represent social benefits beyond direct beneficiaries.

Heterogeneity of Impact: Variation in project impact across different beneficiary subgroups (e.g., by gender, farm size, wealth status, or geographical location). Assessing heterogeneity is important for understanding equity implications and tailoring interventions to different contexts.