YI FSWA Yams for Livelihoods Impact Evaluation Protocols for Agricultural Projects: The Case of Yam Improvement for Income and Food Security in West Africa Djana B. Mignouna, Adebayo A. Akinola, Tahirou Abdoulaye, Arega Alene, Norbert Maroya, Beatrice Aighewi, Thomas Wobill, and Robert Asiedu Livelihood Outcome Y1 (observed) Beneficiaries Y1* (counterfactual) Non-Beneficiaries Y0 t = 0 t = 1 time Benefits = (Y1-Y0) – (Y1*-Y0) YIIFSWA Working Paper Series No. 7 2016 Impact Evaluation Protocols for Agricultural Projects The Case of Yam Improvement for Income and Food Security in West Africa Impact-Evaluation Guidelines Djana B. Mignouna, Adebayo A. Akinola, Tahirou Abdoulaye, Arega Alene, Norbert Maroya, Beatrice Aighewi, Thomas Wobill, and Robert Asiedu YIIFSWA Working Paper Series No. 7 i © IITA 2016 International Institute of Tropical Agriculture PMB 5320 Oyo Road Ibadan, Nigeria www.iita.org ISBN 978-978-8444-67-1 Printed in Nigeria by IITA Correct citation: Djana B. Mignouna, Adebayo A. Akinola, Tahirou Abdoulaye, Arega Alene, Norbert Maroya, Beatrice Aighewi, Thomas Wobill, Robert Asiedu. 2016. YIIFSWA Working Paper Series No 7. The Case of Yam Improvement for Income and Food Security in West Africa, International Institute of Tropical Agriculture, Ibadan, Nigeria. 70 pp. Cover photo: Better harvest had and stored ready for sale in Lambata market, Niger state in Nigeria. ii Contents 1. Project Organizational History 1 Basis of the Project Yam Improvement for Income and Food Security in West Africa 1 The impact pathway for the YIIFSWA project 1 2. An Overview of Impact Evaluation 4 Impact Evaluation 4 Challenges for evaluation 6 How will the project make its impact on poverty? 7 Lessons on putting IE in agriculture into practice 7 Conceptual framework 8 3. Methodological Approach to IE 10 Introduction 10 Design approaches for IE 11 YIIFSWA: Contributory cause and causal packages 15 Designing data collection procedures 22 4. Survey Design, Planning, and Execution 24 Study area 24 Sample size determination 24 Data collection instruments 25 Indicators for assessing project impact 26 Training of enumerators and supervisors 27 Field data collection, data entry, and database management 28 Field data management 29 References 30 iii Tables 1. YIIFSWA processes and characteristics of technology, actors, and community. 9 2. Indicators for Assessing Project’s Impact on Targeted Farmers. 27 Figures 1. A generalized impact pathway of the YIIFSWA project focused on enhancing income and food security. 3 2. Impacts of agricultural innovation, YIIFSWA conceptual framework. 9 3. Levels of impact from plot, individual, household, and community. 10 Annexes 1. YIIFSWA Project Communities in Nigeria. 34 2. YIIFSWA Project Communities in Ghana. 38 3. Household level questionnaire 40 4. Community level questionnaire 62 5. Field level questionnaire 65 iv Project Organizational History 1 Basis of the Project Yam Improvement for Income and Food Security in West Africa Numerous problems have been identified in the yam food sector that impede efforts in national policy programs aimed at promoting yam as a priority crop in the various countries in West Africa. The high cost and unavailability of disease-free seed yam are major challenges which are linked to a competition between seed and food uses (Ironkwe 2005). This situation is compounded by a high incidence of destructive pests and diseases at both pre-harvest and post-harvest stages; the high labor input associated with land preparation, planting, staking, weeding, and harvesting; and the increasing shortage of virgin land (fertile soil) suitable for the production of the crop. These problems are associated with the low production and post-production technologies that are available in the sector (Maroya et al. 2014). As a way of addressing these myriad constraints, a project known as Yam Improvement for Income and Food Security in West Africa (YIIFSWA) was initiated. The project aims in a 10-year horizon to double productivity, stimulate a sustainable increase in incomes for smallholder producers, and contribute to their food security and economic development. The YIIFSWA project is led by the International Institute of Tropical Agriculture (IITA) and funded by the Bill and Melinda Gates Foundation. IITA is working together with various partners, service provider organizations, the private sector, farmers, and traders (Maroya et al. 2014) to find solutions that will reduce poverty and increase food security through investments in the yam sector. The vision for the project’s first five years is to increase by 40% the productivity (yield and net output) of 200,000 smallholder farmers in Ghana and Nigeria, and deliver key global good research products. After an external mid-term review, the initial seven objectives of the project were repackaged into two major components. One is the seed component, which deals with the development of the formal and informal seed yam systems. It focuses on the reduction of postharvest losses, development of technologies for high ratio propagation of high quality pre-basic and basic seed yam, and identification of more effective tools and strategies for the prevention and management of pests and diseases. The second component (leadership, governance, and partnerships) includes project monitoring, evaluation and learning, communication and information dissemination, project coordination and management, as well as the evaluation and scale-out of production technologies using popular new and local varieties. To achieve the project’s vision, an impact pathway was designed identifying the itinerary from which the intended project interventions should achieve the desired impacts. Together with proper problem diagnosis, the impact pathway was critical in determining what interventions were needed, when and where, to achieve the envisioned results. The impact pathway for the YIIFSWA project The impact pathway provides a useful way of conceptualizing the cause-and-effect relationship between different types of changes while impact assessment focuses mainly on the changes occurring at the outcome and impact levels (Maredia 2009). Figure 1 illustrates a simplified impact 1 pathway (or a results chain) to show how actions related to YIIFSWA affect the goal of enhancing income and food security. It also introduces the concept of impact assessment, which evaluates final effects (long-term impacts on poverty, hunger, etc.,) and intermediate effects (medium-term outcomes on production, income, consumption, and prices) caused by the project’s activities (Baker 2000).The development hypothesis was premised on the following theoretical assumptions: (i) Use of appropriate technologies will stimulate increased productivity and production in Nigeria and Ghana; (ii) Linkage of yield and production with the market “pull” along the value chain will reduce transaction costs of production and marketing; (iii) The reduction in transaction costs will ultimately contribute to increased income and food security for smallholder farmers, especially women. In this regard, YIIFSWA supports key innovations through different inputs/activities across its objectives. These include: i) Pest- and disease-free seed yam on a commercially viable basis capable of increasing yield by at least 50%; ii) Post-harvest storage and handling technologies capable of reducing tuber losses by at least 30%; iii) Access to markets enabling smallholder farmers to increase sales and generate needed cash incomes. It is envisaged that the key to stimulating production will be improved accessibility and the use of pest- and disease-free seed yam, coupled with enhanced post-harvest and handling technologies with better access to markets. On the other hand, it is also envisaged that increased access to markets will help to reduce the transaction costs of smallholder farmers and in turn increase profitability and then their income. The model explains how an intervention is expected to lead to intended or observed impacts. A series of expectations and assumptions identify the presumed relationships among the following: y inputs generating various activities; y activities and their immediate outputs/intermediate outcomes at various levels; and y intended effects (such as households and communities that have become financially self- sufficient and food-secure from their own production, lower post-harvest losses, etc.). The favorable impact of new technologies and practices on the lives of farmers in yam-growing areas is an important barometer of the contribution that YIIFSWA will achieve to development and in particular to human resource development. This working paper guides the collection and processing of data and analytical information about the impact of YIIFSWA on households and communities. This is the reference in analyzing trends at community, household, individual, and plot levels as measured and assessed through various surveys. The study is meant to provide answers to the following questions: 1. What are the technologies/practices brought by YIIFSWA in the targeted areas? 2. What are the levels of adoption of these technologies/practices? 3. What has been their impact since introduction? 4. Are there constraints holding back farmers from adopting the new technologies/practices that need attention so they could benefit the broader society? This working paper intends to organize implementation by providing suggestions on designing impact evaluation (IE) for the project on production, productivity, and profitability. The document considers the challenges of conducting an IE of agricultural projects as well as the methods for assessing impact. Issues of collecting agricultural data for an IE and how to put together the design strategy in an evaluation plan are also covered. Moreover, it outlines a broad methodological approach for the 2 process and IE of YIIFSWA. The study is structured into four parts. Section 1 provides the background by discussing the project generally. Section 2 introduces the methodology, discusses the specific challenges of conducting IEs of such agricultural projects, and describes a conceptual framework for the evaluation of YIIFSWA interventions on poverty. The possible methods for assessing impact are presented in Section 3; Section 4 addresses practical issues of data collection. INPUTS BMGF funding, staff & all other kinds of support. Activities Project expectations Outputs/Outcomes Project assumptions Impac t  Functional and reliable pre-basic, basic and certified Sustainably reduce Range of activities covered seed yam units poverty & increase by the project through its  Tuber pest damages in storage barns of participating food security two components covering farmers reduced all objectives and sup-  Farmers using clean planting material increase their ported by cross-cutting yields elements  Standards for high quality clean seed yam production formalized  Farmers linked to the markets and generating in- comes through the increased production and market- ing of ware and seed yams  Yam is affordable by the urban and rural poor con- sumers  Preferred stress tolerant varieties widely grown by farmers  Diagnostic tool kits available for production of clean seed yam and certification  Availability of technologies for high ratio propagation Figure 1. A generalized impact pathway of the YIIFSWA project focused on enhancing income and food security. 3 An Overview of Impact Evaluation 2 This section provides an overview of IE, followed by discussions on its importance as evidence-based policy and later emphasizes the modality of the IE to be used for YIIFSWA. Impact Evaluation The IE is rooted within broader monitoring and evaluation systems and provides a core set of tools that stakeholders can use to focus on results. Borrowing from the OECD-DAC Glossary (2002) as the most widely shared definition, impact is considered as change, positive and negative, primary and secondary, produced by a development intervention, directly or indirectly, intended or unintended. Impact occurs at multiple levels and time frames that can be short-term, intermediate, and long-term changes resulting from an intervention. Impact occurs in different ways depending on the type of intervention and the context. An IE is a systematic and pragmatic study that measures the changes that are attributable to a defined intervention, attempting to establish whether the intervention has made a difference in the lives of people. According to UNEG (2013), the IE enables the process(es) by which impacts are achieved to be better understood and those factors that promote or hinder their achievement to be identified as an important feedback into ongoing or future initiatives, including the adaptation of successful interventions to suit new contexts. However there is reluctance in carrying out IEs because they are deemed to be expensive, time consuming, and technically complex. Moreover, findings from IEs can be politically sensitive, particularly if they are negative. Similarly, many evaluations have been criticized because the results did not come early and did not answer the right questions, or were not carried out with sufficient analytical rigor. The IE is critical especially for developing countries where resources are scarce and the spending of every dollar should have the aim of maximizing its impact on poverty reduction. The lessons to be learnt from IE studies will also provide critical input to the appropriate design of future projects. This working paper seeks to provide general knowledge and the tools needed for evaluating the impact of agricultural projects with a focus on the YIIFSWA project. An IE can use qualitative or quantitative methods or both. Good research that convinces a range of clients on the difficult questions of causality always requires a combination of techniques. For attribution, this section describes the benefit of using a quasi-experimental method to assess the impact of YIIFSWA as compared with non-experimental methods and then discusses the methodological and sampling approaches to be implemented. In the case of YIIFSWA, an IE aims at: (i) estimating the early impacts, positive and negative, primary and secondary, that result from the project; (ii) assessing the direct and indirect contribution of the project on smallholder yam farmers, whether intended or unintended, and (iii) providing lessons that can be learned. The IE provides answers to questions about what goes well or badly, how, for whom, and why. To provide these answers, the IE links cause and effect: it assesses the direct and indirect causal contribution of the project to change in people’s lives. Attribution is the dividing line between an IE and less rigorous forms of evaluation: it is the evidence that the project actually caused the effect 4 being measured. The challenge of attribution is the central problem to be solved in IE (Suresh et al. 2007; Harold 2007). Unfortunately, the only way to truly know whether an intervention caused the observed effect is to compare the effect on individuals who participated in the project with what would have been the effect on those same individuals if they had not participated in the project. This state of non-participation of the participant is known as the counterfactual outcome. The IE requires a credible and rigorously defined counterfactual that estimates what would have happened to the beneficiaries in the absence of the project. The standard challenge in any IE is therefore determining what would have happened in the absence of the project. To truly understand the impact of a project on a given indicator, information would ideally be available on project beneficiaries with the project and those same beneficiaries without the project. The indicator could then be compared between these two states to see if the project had had an impact. To be legitimate, this counterfactual or control group would need to be exactly like the project beneficiaries or treatment group, except that they would not receive the benefits of the project. Thus, any differences in the indicator could be attributed to the project. Creating a counterfactual through identifying a reasonable control group and ensuring that an identified impact can be attributed to a project are always challenges. This document will discuss evaluation methods that allow for attribution. Some key challenges are specifically related to the evaluation of farmer- targeted agricultural projects. Professional evaluators (particularly economists) have developed methods to produce counterfactual estimates since an individual or household or community cannot be both participant and non-participant in the same project at the same time. Counterfactual estimation is usually achieved through the use of a control group. However, for a control group to represent an unbiased counterfactual, the people who do not receive the treatment must share so many characteristics with the treatment group that the two are statistically indistinguishable from each other. Thus, in the rigorous evaluations of development projects, the best control groups are known (and proved) to share many characteristics with the treatment group. Frequently recorded examples of the characteristics that should be common to both groups include the following: location, age, livelihood, level of education, household size, language, tribal-ethnicity, consumption expenditure, access to capital, health status, marital status, sex, and access to public services. To produce an unbiased counterfactual estimate, it is not enough to compare a treatment group with an arbitrarily selected control group (or what is known as a comparison group in non-experimental designs). The control group in an impact assessment must provide evidence of being a reliable counterfactual so it must be statistically capable of actually being the treatment group. Randomized control trials (that are referred to as RCTs or experimental designs), with an appropriate use of mixed methods, are IE methodologies that generally provide the greatest opportunity for learning and constituting a strong counterfactual. However, when an RCT is not feasible or desirable, quasi-experimental IEs that use methodologies such as Difference-in-Difference, Propensity Score Matching, and Regression Discontinuity Design, combined with mixed methods, are other means that facilitate learning and allow for the attribution of impact. The IEs serve two key purposes– providing both accountability and learning. Accountability compares the costs and impacts on final outcomes, such as income and poverty that are attributed to project investments. Learning relates 5 to the development of hypotheses and explores how well or how poorly a particular development approach works. It provides a better understanding of the causal chains expected to link the project investments with income changes. For example, trained yam farmers should: i) learn why positive seed yam selection practices increase yields; ii) adopt these practices; iii) improve their yields; iv) increase farm income; and v) ultimately raise their household incomes. Learning requires understanding how and why these causal linkages do or do not happen and why it is essential to test the assumptions behind the project design. The IEs are an essential tool for learning and for accountability, but they are not the right tool for every project. They should be used selectively, with a special focus on where the potential for learning is greatest. Evaluation in the agriculture sector is especially challenging due to several factors that fall into two broad categories–how agricultural projects can pose challenges for evaluation and how evaluation approaches can cause challenges for the implementation of agricultural projects. This working paper will not discuss these combined factors which can make implementers and sector specialists hesitant about rigorous IE. The concern will focus on the main challenges related to the yam sector and attempt to develop practical solutions for managing IEs in this context. Challenges for evaluation Several factors such as the crop cycle and weather variations are key factors for evaluation related to yam. Others, such as self-selection, are relevant to a variety of sectors but can be especially magnified in agriculture in general. Yam cycle and seasonality The seasonality of agriculture creates two challenges related to timing. The first is relative to the yam cycle. Yam are mainly considered an annual crop and the life cycle poses strict windows for when training and related activities can occur. If these windows are missed because of delays in project implementation or evaluation planning, a full crop cycle can be lost. This has implications for the project’s ability to achieve its objectives and to evaluate impact. The second is the expected time between an intervention and the expected results. Agricultural projects often require several crop cycles to yield benefit, as farmers become proficient in new techniques, expand their application, and learn from one season to the next. In addition, with some projects, the difference in outcomes between farmers in the treatment and control groups should substantially grow over time. This creates challenges for evaluation when this is done very soon after a project has been completed because impact can be seriously underestimated. Context variables and risks Yam production is severely affected by natural disasters and other unknown and unpredictable phenomena; conditions which are susceptible to change from time to time. In theory, a valid control group should, on average, face the same weather shocks as the treatment group, but interventions may influence the magnitude of weather effects on outcomes, implying that weather shocks can influence impact estimates. For instance, at Idah in Kogi State and Illushi in Edo State, Nigeria, there was a severe flood along the banks of the river Niger and farmers’ yields were affected in years 1 and 4 of the project. Hardly any results were achieved those years because most of the demonstration plots that had been set up and the planting material were destroyed by the flood. 6 In Ghana, an erratic rainfall pattern and a long drought in the Northern region have affected some project activities as well as occasional attacks by cattle on yam fields in both Ghana (Ashanti and Northern regions) and Nigeria (Nasarawa and Niger States). Spillover effects Spillover or demonstration effects are sometimes to be expected, such as when people outside the primary targeted beneficiary groups adopt techniques supported and promoted by YIIFSWA to gain from a desired outcome. If YIIFSWA indirectly affects the outcomes of the control group even though the control group itself did not participate in the activities, there will be biased estimates of impact. Spillover effects can be quite large in this project where technology is easily transferred and are often an explicit component of project logic. Implementation changes Even when the project design is set, its implementation approaches may require significant changes over time in response to project goals and mid-term evaluation. Adjusting implementation approaches makes interventions more effective and improves beneficiaries’ targeting. However, this could challenge the validity of the evaluations and reduce the potential for learning what really works. How will the project make its impact on poverty? Assets are the primary transmission channel by which YIIFSWA expects to make an impact on poverty. Specifically, it is envisaged that the skills and knowledge of project clients will be strengthened through training and demonstrations. In some cases, this knowledge may be translated into increased physical capital. More fundamentally, it is anticipated that building human capital will catalyze increases in financial capital through higher farm revenues as a result of the uptake of new technologies. It is also expected that certain outcomes will result in enhanced human capital in the form of improved nutritional status among members of client households. Participation in group activities to promote technology dissemination might promote enhanced social capital in some instances, particularly for women, if their subsequent successful adoption of the promoted technologies is the result of their participation, but this is by no means a primary outcome. Employment might also serve as a secondary transmission channel for poverty reduction if production gains for commercial operations are of sufficient magnitude to stimulate expansion warranting the employment of additional hired labor, and if efforts to strengthen input supply-side value chains create significant additional demand for related services. Lessons on putting IE in agriculture into practice Although these challenges are real and can be difficult to manage, they should not prevent the pursuit of rigorous IE in the agricultural sector. Given the critical role of agriculture for development, and the tightening of development budgets worldwide, it is essential that the development community should deepen its understanding about what approaches work best to reach desired outcomes in a cost-effective way. Many of the lessons presented identify approaches that facilitate the use of a counterfactual, which compares the changes that occur both with and without a given intervention through the use of treatment and control groups. These solutions also adhere to evaluation methodologies that maintain the integrity of counterfactuals so that IEs can identify attributable impacts. 7 The lessons are designed to support donors, partners’ countries, implementers, and evaluators in striking a good balance between achieving impact, measuring results, and learning what works in agricultural investments. The following challenges and lessons emerge from discussions about IE in the yam sector but many are also broadly applicable to other developmental sectors. i. Define early the project logic and objectives of the evaluation, and how to integrate the two. The most important first step - both for successful implementation and evaluation - is to have a clear picture of what a project aims to achieve and how planned interventions are expected to lead to that outcome. ii. Engage early and communicate often. Coordinated planning and ongoing communication are essential ingredients for minimizing and managing trade-offs between implementation approaches and evaluation methodologies. iii. Foster joint ownership by aligning incentives. Everyone involved must feel ownership over both the implementation and evaluation of the project, so incentives must be aligned for donors/ sponsors, partners’ countries, project implementers, and evaluators. iv. Match evaluation methodology and project design. The most rigorous method for measuring attributable project impacts is through RCTs, but when they are not feasible, there are other rigorous methods to be considered for evaluation. v. Focus on long-term impacts but be prepared to show early results. The IEs are often not carried out for a year or two after project completion. While planning to be accountable for progress and to communicate early results, the measurement of long-term impact must not be neglected. Conceptual framework The conceptual framework for this methodology was developed for interventions which promote technological innovations such as the adaptive yam minisett technique (AYMT), vine propagation, conventional tissue culture, aeroponics and bioreactors systems, diagnostic tools, new and existing technological packages for ware yam production, also varieties for adaptation to environments with low soil fertility and low moisture stress, as well as labor‐saving systems, and crop management and postharvest practices. In this project, clients (technology user groups) and service providers (research and extension agencies) work together on the adaptation and uptake by clients of particular technical or institutional innovations, The first stage in developing a methodology to assess innovations’ impacts on the poor is to identify (i) the different processes by which innovations affect the poor; and (ii) the factors affecting these processes. Figure 2 shows a simple schema of four elements, steps, or processes (elaborated in Table 1) by which YIIFSWA can exert an impact on different members of a rural community. The project undertakes a range of experimentation, adaptation, capacity building, and organizational development activities (1). These lead to a process of innovation adoption by the target audience and by others in the community (2). Adoption then results in ‘direct impacts’ on the livelihoods of these adopters (in this case by increasing their incomes or improving food security) (3). Changes in the productive activities and livelihoods of these adopters will then have indirect positive and/or negative impacts on non-adopters (4). 8 Clients YIIFSWA Processes of innovation adoptio n Project (2) (1) 1 Service providers (+/-) Characteristics of the product and innovation- Characteristics of the acto rs 2 3 (+/-) Direct impact on adopters Indirect impact (3) (4) Figure 2. Impacts of agricultural innovation, YIIFSWA conceptual framework. (Adapted from Paz et al. 2006) Table 1. YIIFSWA processes and characteristics of technology, actors, and community. Processes Characteristics of the Characteristics of the actors Characteristics of the technology/practice community Innovation Relative advantage, Objectives pursued and livelihoods Networks, local organizations, adoption complexity, compatibility, (pathways, existing activities), associations, demographics, feasibility, observable results, network membership, knowledge, (population density, a connection between direct education, gender, risk aversion, distribution, age structure), impact and the options need for and availability of roads, telecommunications, and objectives pursued, resources, perceptions regarding gender relations, other beliefs services/institutions required, direct and indirect impacts, and norms, educational perceptions regarding direct power relations, ability to link the facilities, other services, past and indirect impact, etc. requirements of YIIFSWA project experience, price tendencies, with local organizations, etc. natural and local resources, etc. Direct impacts Profitability, productivity, quality, Livelihoods, activities, roles and Local organizations, access uses (commercialization purposes of the activities, capital to markets, labor market or subsistence), resource (financial, social, natural, physical integration, roads, presence requirements (labor, capital, and human) holding and access, of innovators, other services, land, skill (type/quality)), vulnerability, other risks and telecommunications, prices variability, prices (qualitative uncertainties, social relations and trends and their vulnerability, evaluation of sensitivity roles, gender, etc. alternative opportunities, for investment/ cash and economic growth / stagnation credit flow), services / (by social levels), etc. institutions required, livelihood contributions, goods tradability, use of surplus, etc. Indirect Requirements for skilled / Adopters and their relation with Local market size, labor impacts unskilled labor and for inputs the labor market, investment market integration, elasticity and services for production, and consumption patterns (good and structure of the inputs for processing; tradability) labor market, investment profitability (for reinvestment opportunities, local supply of and consumption expenditure); goods, income distribution, product tradability, etc. general consumption patterns (tradability), etc. 9 Methodological Approach to Impact 3 Evaluation Introduction Indicators to be employed for impact assessment are the yields of R&D plots relative to yields of farmers’ non-demonstration plots, an increase in the production of yam in project areas, yam value- added products, coverage of use of new practices in project areas, total increase in income and food security, training of farmers, NGOs, entrepreneurs, extension and research officers, and other stakeholders depending upon locations. However, the process of impact assessment will be done at different levels but predominantly at farmers’ and households’ levels as illustrated in Figure 3. The results framework for YIIFSWA reflects a commitment to embark on rigorous and systematic methods of projecting, tracking, and evaluating its impacts. Together with transparency, this approach is a cornerstone of YIIFSWA’s obligations to accountability and learning where M&E is committed to making its evaluations as rigorous as warranted to understand the causal impacts of the project on the expected outcomes. Evaluations support two objectives derived from these core principles as mentioned earlier: accountability and learning. Accountability refers to the project’s obligations to report on its activities and attributable outcomes, accept its responsibility, and disclose its findings in a public and transparent manner. Learning refers to improving the understanding of the causal relationships between interventions and changes in poverty and incomes. Community level - Farmers’ participation - Social networks - Employment Household level - Food security - Income/Expenditure - Assets - Gender Individual level - Skill development - Well-being - Leverage in decision making - Control of resources - Specific participation Plot level - Area - Output - Resource management Figure 3. Levels of impact from plot, individual, household, and community. 10 No single evaluation methodology can respond to the questions of interest to clients and stakeholders about many kinds of evaluation, nor can any single design understand and respond to interventions from YIIFSWA. Therefore, the study will use a variety of methods and approaches to assess the impacts of the project. Prospective evaluations were developed at the same time that the project was designed, and were built into project implementation. Baseline data were collected prior to implementation for both treatment and control groups. Prospective IEs were adopted to produce strong and credible evaluation results, with the generation of baseline data to establish pre-project measures of outcomes of interest. This provided advance information on beneficiaries and comparison groups. The baseline survey served as a foundation for a before/after comparison of pre‐ and post-treatment states. It therefore allows for the application of a quasi-experimental design, which is discussed in the next section, together with the qualitative means to be used to collect data. Design approaches for IE On the basis of the review of methodological literature and of existing IE cases that used a broader range of methods, the following designs and methods were chosen. Experimental and statistical approaches Economists have increasingly emphasized the use of RCTs to determine the effectiveness of development assistance (Duflo and Kremer 2005; Banerjee 2007). Although there is some debate over whether RCTs are the only valid approach (Ravallion 2009), there seems to be agreement on the value of carefully collecting data to evaluate the impact of development projects and the importance of using carefully constructed datasets and empirical approaches to identify impact. Agricultural evaluations are often complicated by indirect or “spillover” effects that are due to the transferring of new technologies and management practices from project participants to non- participants. In fact, agricultural interventions, particularly for technology adoption projects, often explicitly aim at facilitating spillover effects. Although these factors often increase the impact of the operations, they complicate the evaluation design by making it hard to find an “uncontaminated” counterfactual. Additionally, since much of the influence of the project may be through these spillover effects, a correct assessment of the project requires considering how to identify them when they exist and to ensure a reasonable, uncontaminated counterfactual (Angelucci and De Giorgi 2009; Angelucci and Di Maro 2010). In this study, attempts would be made to properly document and capture the spillovers of the intervening technologies of YIIFSWA project. Although the data and evaluation design issues require special consideration, evaluating agricultural projects can be done using standard approaches with some modifications. The purpose of this guideline is to provide suggestions on designing IE for agricultural projects. In particular, the working paper focuses on YIIFSWA which directly targets smallholder farmers, and seeks to improve production, productivity, and profitability. A number of approaches can be taken in evaluating projects. There is a range of accepted approaches to determining an appropriate comparison group for counterfactual analysis, using either prospective (ex-ante) or retrospective (ex-post) evaluation design. Prospective evaluations 11 begin during the design phase of the intervention, involving the collection of baseline and end- line data from intervention beneficiaries and non-beneficiaries; they may involve the selection of individuals or communities into treatment and comparison groups. Retrospective evaluations are usually conducted after the implementation phase and may exploit existing survey data, although the best evaluations will collect data as close to the baseline as possible, to ensure the intervention and comparison groups are comparable.. The IE designs are identified by the type of methods used to generate the counterfactual and can be broadly classified into three categories, the experimental, quasi-experimental, and non- experimental, that vary in feasibility, cost, and involvement during the design phase or after the implementation phase of the intervention, and the degree of selection bias. Since data collection tends to be representative samples of treated and control households or individuals, statistical methods, particularly coming from the econometrics literature, are used to identify impact. The best method for assessing impact for a given project depends on the data available. As a general rule, the better the data, the less sophisticated the econometric procedures that are needed for analysis. Because such models necessarily rely on certain assumptions, there has been an increasing emphasis on collecting better data to avoid having to use more complicated econometric procedures. For YIIFSWA, the quasi-experimental design will be used. Quasi-experimental design To assess impact, it is necessary to identify a counterfactual and then to take measures to ensure the estimate of impact is free from bias. Quasi-experimental methods include matching, differencing, instrumental variables, and the pipeline approach. If selection characteristics are known and observed, they can be controlled to remove the bias. Matching involves comparing project participants with non-participants, based on observed selection characteristics. Propensity Score Matching (PSM) uses a statistical model to calculate the probability of participating on the basis of a set of observable characteristics and matches participants and non-participants with similar probability scores. Difference-in-differences or Double Differences, which use data collected at baseline and end-line for intervention and comparison groups, can be used to account for selection bias with the assumption that unobservable factors determining selection are fixed over time (‘time invariant’). Estimation of instrumental variables accounts for selection bias by modeling participation using factors (‘instruments’) that are correlated with selection but not the outcome, thus isolating the aspects of program participation which can be treated as exogenous. One common issue with evaluating agricultural projects is that they often involve self-selection of participants. For example, agricultural extension projects usually interact with self-formed groups of farmers. Self-selection implies that only farmers of certain types may choose to participate in a given project. If an evaluation attempts to determine the impact of a project by comparing those that chose to be in the project with those that did not, the differences in the indicator of interest may reflect not only the impact of the project but also any innate differences between participants and non-participants. Suppose the better yam farmers in a region decide to participate in an agricultural extension project – that is, farmers who are innovative and like to experiment with their production to see what works best. Such yam farmers are likely to have higher yields even without the project. A comparison of yields between these innovative, treated farmers and non-participant, control farmers 12 is likely to show higher yields for the treated farmers due to the project but also due to the fact the farmers are innovative. The problem is that it is hard to know how much of the yield difference is due to the project and how much to the differences in the types of farmer. This makes any estimate of project impact biased since the estimate cannot solely be attributed to the project. Clearly, selection is also an issue if farmers with certain attributes are chosen by the project to participate. If a project focuses on farmers with limited land access, those with larger landholding are unlikely to provide a good comparison. However, these attributes tend to be observable since the project must observe them to identify who will participate. With YIIFSWA, a careful evaluation design was done in combination with project design before implementation, creating a reasonable counterfactual and avoiding biased estimates of impact. Self-selection can also be managed, as discussed in the next section, but tends to be more complicated. In YIIFSWA we expect a strong evaluation function and feedback loop that will enable us to be accountable in both cases, and to learn from each so that we can make continuous improvements. We can do this only with evidence and data to inform our decisions. The evaluation in YIIFSWA will mainly use the DD approach and PSM.. Difference-in-Differences approach The DD approach is one of the most popular non-experimental techniques in IE since it allows controlling for some types of selection in a straightforward and intuitive way, as long as baseline data are available. In a DD model, the relevant comparison is the changes in the indicator over time. Here, the difference of outcome indicator levels is measured for both the treatment group and a control group, before and after the treatment. First, for each group the mean difference between the outcome indicator levels in the pre‐ and post‐intervention periods is calculated. The difference between these two mean differences is subsequently calculated. This two‐step approach gives the method its name1. The impact of the project is thus defined as: (Yt ' −Yt | D = 1)− (Yt ' −Yt | D = 0) With: t being the time of the baseline and; t’ the time of the post‐treatment survey. The result equals the project’s impact if the underlying assumption holds true that the difference between before and after the intervention in the control group can serve as a proper counterfactual for the treatment group (Wooldridge 2001). The end-line surveys, necessary for calculating the impact estimators, should be as comparable to the baseline survey as possible, ideally encompassing the same survey design, same questionnaire, same interviewers, etc. It would be best even to ask the same respondents but if this is not feasible, going for the same geographic clusters or strata is recommended, especially for some other variable(s) (Baker 2000). 4 The approach is named non-uniformly in the literature, the most common terms being Double-Difference- method, or otherwise Difference-in-Difference estimator. 13 The great benefit of using DD is that it controls for unobservable differences in the baseline characteristics of treatment and control households, thus minimizing potential biases in impact estimates. The DD estimates address only time-invariant differences in control and treatment groups. This means that if there are changes that occur over time that affect one group and not the other, this cannot be controlled by using this approach. The evaluator must be confident that such changes did not occur to be sure that impact estimates are reasonable. The most widely applied method for evaluating donor interventions is probably DD, but it is also the most data-intensive. (See Duflo 2001, DiTella and Schargrodsky 2005, and Todd 2008 for a review of studies applying DD estimators.) The method as already outlined above relies on comparing the outcomes of interest for the treatment group with a control group both before and after the intervention. Therefore, the applicability of the method crucially depends on the availability of the baseline data. This approach will be used in conjunction with PSM, where DD will be estimated for the matched groups. Propensity Score Matching approach This approach is based on the selection of a group most similar to the treatment group in terms of the probability of being selected which is derived from accumulated contributions from observed characteristics. Economic impacts will be assessed using PSM to control for the self-selection into adoption that normally arises when technology adoption is not randomly assigned. The main parameter of interest in a non-experimental framework is the Average Treatment effect for the Treated population (ATT), expressed as: Where: Y1 denotes the value of the outcome when the household adopts the technology (1), and Y0 is the value of the same variable when the household does not adopt (0). The problem that arises with unobservability is by virtue of the fact that) can be estimated but not. Although τ= can normally be estimated, it is potentially a biased estimator ofτATT . To ensure the reliability of PSM, participants and controls have the same distributions of unobserved characteristics. Failure in this condition is often referred to as a problem of “selection bias” in econometrics, or “selection on unobservables” (Heckman and Robb 1985). Also, the support for the comparison and the program participants should be the same. Finally, the same questionnaire will be administered to both groups and participants and controls must be derived from the same economic environment. Other approaches Other approaches exist and are not currently widely deployed in IE. They will be used since they offer considerable potential for linking interventions with outcomes and impacts. Some of these approaches are discussed below. 14 Case-based approaches These approaches might be case-studies or outside traditional acceptance as case-studies. They may be policy interventions, institutions, individuals, events, trainings, or demonstrations during a particular period. This represents a shift from focusing causal analysis on variables taken out of their specific context. Locating variables in the context of the ‘case’ and conducting within-case analysis alongside comparisons across cases has opened up major new opportunities for causal analysis that are still largely ignored in evaluation practice. The design will take into account the case studies that generally focus on the unique characteristics of a single case. These case studies avoid causal analysis even though they contribute to such analysis in several ways. For example, interpretative case studies help to define construct validity in terms that make sense to stakeholders on the ground and give voice to project beneficiaries, both at the stage that evaluation questions are formulated and when interpretations of findings are being made. These approaches tend to generalize under certain conditions and identify clusters or subsets of cases about which it is possible to make similar causal inferences. Participatory Approaches Development evaluation has expressed ideas such as participation in ownership through methods such as Participatory Rapid Assessment and Participatory Action Research, These approaches relate to IE, even if only indirectly. They could help in the following: y Ensuring that beneficiaries have a voice from the beginning of YIIFSWA, thus improving its plans and interventions. y Investigating local communities and circumstances, clarifying problems and constraints, thus improving ‘construct validity’. y Adding a beneficiary and stakeholder perspective to the conclusions and lessons learned from an IE. This last point raises the question: who participates? In YIIFSWA, those participating include beneficiaries but also country-based officials and decision-makers. There will be different implications from different patterns of participation. For example, the participation of decision- makers may have implications for efficiency and sustainability in implementation. To be seen as a design that contributes more directly to IE, participatory approaches need to support causal inference. Participatory approaches to causal inference do not see recipients of aid as passive recipients but rather as active ‘agents’. Within this understanding, beneficiaries have ‘agency’ and can help ‘cause’ successful outcomes by their own actions and decisions. As suggested above, this is the case of YIIFSWA where country-based decision-makers were actively involved in the project and its evaluation. It should be possible to interpret impacts in terms of a participatory content for interventions: for example, the extent to which involvement, ownership, and commitment improve development outcomes. This is being explored in this study. YIIFSWA: Contributory cause and causal packages Simple sufficient causation could be more promising in that an intervention on its own may be sufficient to produce the impact but in YIIFSWA, many interventions are seen as a ‘contributory’ cause and are demanding conditions for impact to occur. There are a variety of ways that such 15 impacts might be realized, for example, quality training outcomes and empowerment. Contributory cause in this case recognizes that effects are produced by several causes, none of which might be necessary or sufficient for impact. It is difficult for statistical and econometric models to deal with multiple causalities and to capture the influence of combinations of causal factors rather than of each factor as a free-standing agent. The causal package consists of the delivery mechanism for a variety of agricultural products and services such as input development, distribution, trainings, and demonstrations. These products and services include: (i) Training in minisett and vine cutting technologies; (ii) Training in business plan development for pre-basic and basic seed producers; (iii) Training in business plan development for seed producers; (iv) Training in business plan development for yam producers; (v) Provision of seed tubers; (vi) Provision of QDS/pre-basic/basic materials for seed production; (vii) Provision of plantlets; (viii) Training on high-ratio seed yam propagation techniques; (ix) Production of certified seed yam; (x) Training in seed yam quality control and certification; (xi) Improved yam storage facilities. To measure the early impact of complex development projects such as YIIFSWA all the agricultural interventions will be assessed as a causal package. However, this evaluation will not give a clean estimation of the effect of a particular intervention. Instead, it will measure the effect of YIIFSWA’s package of interventions as delivered, which will be compared with the status quo of services. Agricultural innovations in YIIFSWA can be classified according to their impact as new products, yield-increasing and cost-reducing innovations, as well as innovations that enhance product quality. Most of the interventions in the work plan reaching farmers do not introduce novel technologies, but rather build upon the existing practices of clients through initiating simple improved management to increase yields. These technologies are also largely appropriate within the context of social and cultural norms regarding gender roles. In this context, attention would be on the role of the AYMT in that package. Was it a necessary ground-preparing cause, a necessary triggering cause, or something that did not make any difference? Would a similar effect have occurred without the intervention? If the intervention was indeed a trigger, then a stronger claim becomes possible. If the intervention starts the causal chain and possibly supports change along the way, it is possible to claim that it was the intervention that made the difference because it was an initiating contributory cause. Adaptive yam minisett technology adoption The availability and affordability of high quality seed yam are among the most important challenges facing the yam sector in West Africa. Seed yam are expensive (Ironkwe 2005), accounting sometimes for as much as 63% of total variable production costs. The multiplication ratio of tubers is very low (less than 1:10) compared, for instance, with some cereals (1:300).This leads to a scarcity of seed tubers which often results in mounds prepared in farmers’ fields remaining unplanted (Aighewi et al. 2002); some farmers also keep a reserve batch of seed yam (up to one-third of the quantity planted) to replace those that do not germinate. This situation has been aggravated by the poor quality of the seed yam, as those that germinate tend to carry problems (viruses, fungi, nematodes, and insects) from the storage barns to the field, thereby resulting in low tuber yields and poor shelf life (Ampofo et al. 2010). 16 To overcome the shortcomings of the traditional methods of producing seed yam in West Africa, the NRCRI and IITA through research efforts developed in 1982 an effective and affordable technique, the yam minisett technique (YMT), for farmers to produce their own seed yam (IITA 1985). With this technique, the multiplication ratio could be increases from the traditional 1:5 to 1:30 (Orkwor et al. 2000). The development and introduction of YMT are key strategies for transforming the sector and for enhancing the well-being of the rural population in West Africa. The technology has been promoted for three decades. However, these efforts have not been evaluated rigorously and, in particular, there is a lack of panel data which could be utilized to empirically trace adoption since the 1980s. Moreover, several studies which have attempted to address the areas (Ironkwe et al. 2007; Bolarinwa and Oladeji 2009; Wiredu et al. 2012; Abubakar et al. 2015) revealed that few households have adopted the new technology and many “disadopted”. Recently not much is heard regarding YMT because it is not being actively promoted and evidently convincing (Aighewi et al. 2002). This technology has not been adopted by farmers, and both adoption and disadoption have been going on simultaneously. Such a challenge has been investigated and this provided an opportunity for YIIFSWA to address the gap on disadoption rates and an AYMT was introduced to strengthen the yam seed system for quantity and quality assurance in both Ghana and Nigeria. On this note, YIIFSWA has been vigorously promoting the adoption of AYMT since its inception in 2011. However, the current level of adoption and its associated impact on farming households are yet to be empirically investigated. Among others, this study would provide this empirical evidence. Supporting and facilitating farmers in their engagement in agriculture are critical keys to improving their welfare through providing information, skills, and technologies. Building capacity can be organized in a variety of forms to increase farmers’ productivity and income. According to Anderson and Feder (2003), productivity increases are possible only when there is a gap between actual and potential productivity. They suggest that gaps of two types contribute to the productivity differential – the technology gap and the management gap. Education can contribute to a reduction of the productivity differential by increasing the speed of technology transfer and by increasing farmers’ knowledge and assisting them in improving farm management practices (Feder et al. 2004). Additionally, it also plays an important role in improving the information flow from farmers to scientists (Anderson 2007). A range of approaches aiming at building farmers’ knowledge have been promoted over the years. A number of models have been implemented since the 1970s, combining approaches to outreach services and adult education, including the World Bank’s Training and Visit model (Anderson et al. 2006), participatory approaches (Hagmann et al. 1999), and farmer field schools (FFS) most recently (van den Berg and Jiggins 2007). Additional modalities include ICT-based delivery which provides advice to farmers on-line and other approaches such as the promotion of model farms (Birner et al. 2006). The YIIFSWA project set up a scheme using participatory approaches with an integrated Training and Visit model to encourage smallholder farmers to produce good quality seeds as well as providing links to retailers of farm inputs to ensure the quality of their produce. Organizations2 in Nigeria were the Missionary Sisters of the Holy Rosary (MSHR), Justice for Peace and Development (JDPM), Arimatheas Foundation for Development (AFD) and Umuasua-Isuikwuato Smallholder 2 NGO based in Nigeria with significant role as service provider in the seed yam systems as well as the dissemination of project outputs and capacity strengthening of farmers in the targeted areas in Nigeria. 17 Oil Palm Farmers’ Cooperative Society Ltd (SHOP) and in Ghana3 the Catholic Relief Services (CRS), Ecumenical Association for Sustainable Agricultural and Rural Development (ECASARD), and Sustenance Ago Ventures and SKY-3 Farms (SAVE). Farmers were grouped in selected communities and assigned to demonstration plots under special supervision and guidance. They benefited from trainings on AYMT; hence, each group treated their setts before planting. Some of the farmers were also trained in business plans with ongoing input demonstrations. Participation is hypothesized to affect the adoption and economic impact of technologies by improving the relevance and appropriateness of the technology to the potential beneficiaries, thereby enlarging the pool of potential adopters. The FFS aimed at providing agricultural technologies for livelihood support. The farmers participated in the schools and learned new agricultural technologies and practices, such as farm management, seedbed preparation, proper spacing, new varieties, and planting techniques. Participants were both males and females and the numbers varied in Nigeria and Ghana. Participants attended lectures on agricultural technologies in (open-air) classes and training at demonstration plots included operations from seedbed preparation to harvest and storage. Sites were purposely selected to deal with “Doubting Thomases”, showing how the setts could perform, yielding good quality seed yam which would be used to produce a great harvest of ware yam. The fields were planted with setts treated with a cocktail of insecticide (Chlorpyrifos, 48EC) and fungicide (Mancozeb, 80WP), at the rate of 100 ml of the insecticide formulation and 100 g of the fungicide formulation per 10 liters of water. In some of the plots, a few rows were planted with untreated setts so that farmers could clearly see the difference between the two. Each plot was farmer-owned and farmer-managed. No problems were reported with the agronomic practices as the inputs were supplied and all farmers also received training on AYMT and other sound practices such as crop rotation. The idea of keeping records of participants is new and somehow creates the expectation of some future help, no matter how great or small. Each site is clearly marked with its own small board which helps those interested to see how this plot is different from their own. Empirical investigation into adoption of adaptive yam minisett technique An important step in assessing the impacts of AYMT is to document its adoption rates. Adoption and the economic impact of AYMT– defined as technology impacts – will be assessed using conventional adoption studies and econometric analysis, complemented by qualitative data from interviews with farmers. The adoption profiles of technologies developed/promoted over time could be derived using the S-shaped logistic function (Griliches 1957), which has been used widely to analyze adoption patterns over time (Feder et al. 1985; CIMMYT 1993; Maredia et al. 2000; Bantilan et al. 2005). The size of the impact of AYMT depends on whether and to what extent – in terms of area planted, for example – the technique has been taken up and grown by farmers. The adoption of AYMT can help to increase productivity, farm incomes, and food security, and so reduce poverty levels, thus improving household welfare. The decision of whether or not to adopt 3 NGO based in Ghana, sub-grantee to IITA under YIIFSWA, in charge of developing and promoting technologies for enhanced seed yam production and development of seed growers. These two key activities seek to contribute to the establishment of sustainable availability of high quality seed yam on a commercially viable basis in targeted areas in Ghana. 18 AYMT hinges upon a careful evaluation of a large number of technical, institutional, and socio- economic factors. The observed adoption choice of AYMT is hypothesized to be the result of a complex set of inter-technology preference comparisons made by farmers. It is common to examine factors affecting the adoption and intensity of use of AYMT by estimating Probit or Logit models of the above-mentioned variables on areas planted with AYMT. This area will thus represent a censored distribution since some farmers (non-users) will assume a value of zero for not adopting. Theoretical model and empirical specifications Adoption is conventionally conceptualized to be the mental process through which an individual passes from first learning about an agricultural innovation to finally adopting it (Mutandwa et al. 2007). In modeling the utility or satisfaction derived from the use and integration of AYMT into the smallholder farming system, the economic values or benefits associated with the technique need to be considered. A typical smallholder farming-household seeks to maximize a multi-dimensional objective function, including increasing incomes and food security and reducing all forms of risk (Strauss et al. 1989). When there is a change in the economic parameters associated with AYMT, the central question is related to how much compensation, whether paid or received, would render the decision-maker indifferent to the change. Thus the change in welfare associated with this development was used as the basis for the economic valuation process. When an individual farmer faces a change in a measurable attribute, for example, higher yield in terms of quality seed yam produced from AYMT (q), then q changes from q0 to q1 (with q1>q0). The indirect utility function u after the change becomes higher than the status quo. Now the status quo can be represented econometrically as follows: u1j = ui (yi,zj, q 0 , ε0j ) On the other hand, the changed or final state due to the introduction of AYMT is shown by: u2j = ui (yi, zj, q 1 , εij ) Where: yi, refers to the farmer’s income, Zj is a vector of the farmer’s socio-economic variables and attributes of choice, and εj is the stochastic error term representing other unobserved utility components. The farmer would opt, pay for, and adopt AYMT in the following conditions: ui (yi – Pi, z j, εij ) >u0 (yi, zj, ε0j ) Where: Pi is the monetary investment associated with AYMT. Since the random components of the preferences are not known with certainty it is possible to make only probabilistic statements about expected outcomes. Thus, the decision by the farmer to adopt AYMT is the probability that he/she will be better off if this technology is used. This is represented as follows: 19 Prob (Yesi) = Prob [ui (yi – Pi, zj, εij )> u 0 (yi, zj, εij )] Since the above utility functions are expressed generally, it becomes critical to specify the utility function as additively separable in deterministic and stochastic preferences. Using, this argument, the function becomes: ui (yi , zj, εij ) = ui (yi, zj ) + εij Where: The first part of the right-hand side is the deterministic part and the second is the stochastic part. The assumptions that εij are independently and identically distributed with mean zero describe most widely used distributions. Determinants of adoption of AYMT Two widely used distributions are the normal (Probit) and logistic regression models (Logit). In this study, the statistical dichotomous choice data are modeled by superimposing a probability function. The dependent variable takes the value 1 if the smallholder farming-households adopt AYMT or 0 if they do not adopt. And if the farming households adopt, how much could they adopt? The observed adoption of AYMT is hypothesized to be the end-result of combined effects of a number of factors related to the farmer’s goals and means of achieving them. The Probit (the standard cumulative distribution function) and the Logit models (Polson and Spencer 1991) will be used for this study. Following Polson and Spencer (1991) and Adesina and Zinnah (1993) the Probit model is: Wi Pr ob(Yesi ) = F (W ) 1 i = ∫ exp(−S 2 / 2)ds −∞ 2π For −∞ < wi < ∞;wi = X i 'β Where: Prob (Yesi) is the probability that the ith farmer chooses to use AYMT, zero otherwise. X is the n by k matrix of the explanatory variables and Beta is a k by 1 vector of parameters to be estimated. The logistic distribution function is closely associated with the standard normal cumulative function of the Probit model. The change in the probability that the farmer uses a purchased input, given change in any one of the explanatory variables, can be computed as: ∂Pr ob(Yesi ) ∂ ∂w = ( F )( i ) = F (w x w x i )β∂ i ∂ i ∂ i 20 Where: F(wi) is the standard normal density (logistic density) function for the Probit (Logit) model. To avoid the censoring bias that Ordinary Linear Square could generate, the Tobit regression model could also be applied to investigate the determinant factors where the ratio of land with AYMT was used as a dependent variable. The Tobit model, originally developed by James Tobin (1958) the Nobel laureate economist (Gujarati 2004), has been useful in several empirical applications in statistics, econometrics, and the adoption literature (Amemiya 1973; McDonald and Moffit 1980; Shakya and Flinn 1985; Adesina and Zinnah 1993; Oladele 2005; Akpoko 2007). The function is estimated from censored samples where the dependent variables have mass points at the low end called limit values and continuous values above the limit. The Tobit model will be appropriate in this study since the dependent variable is the share of land under AYMT; thus the dependent variable must be between 0 limit, and continuous levels of adoption above the limit. A Tobit model censored at zero could be used because a share of land under AYMT smaller than zero will not be observed and some respondents may report a zero share of land under AYMT. The application of this type of limited dependent variable model is not new. A few recent examples include Doss and Morris (2001), Ransom et al. (2003), Nkamleu (2004), and Nkamleu and Tsafack (2007). While other estimation approaches, such as the Heckman’s model, could also generate unbiased results, the Tobit approach conserved degrees of freedom and is relevant in cases such as this one, where the independent variables had a continuous effect on the dependent variable. Generally, the Tobit model uses the Maximum Likelihood Estimation method to estimate the parameters assuming normality and homoskedasticity conditions. According to Greene (2003), the general formulation of the censored regression (Tobit) is an index function shown below: Yi* = β’Xi + εi, Yi = yi* If yi*>0 Yi = 0 if yi* ≤ 0 Where: The index variable, Yi* defines an underlying unobservable tendency where the adoption is a choice rather than a technical outcome. βXi is a vector of unknown parameters and εiis a random error term. The equation above means that the adoption (yi) of AYMT will be observed only when the latent tendency is above the unobservable threshold (yi*>0). If yi* is less than or equal to zero, then yi becomes zero, meaning that there is no adoption. To estimate the probability and the level of adoption of AYMT, the Tobit model using the STATA computer package will be applied on the equation above. The dependent variable Yi i.e., the adoption of AYMT, will be expected to give a value ranging between 0 and 1, signifying that a certain proportion of area is planted with AYMT. The model combines aspects of the binomial Probit for distinction of Yi = 0 versus Yi> 0 and the regression model for E [Yi | Yi> 1, Xi] 21 Where: Y = the proportion of area cropped with yam under AYMT β = vector of parameters to be estimated; and εi= error term Designing data collection procedures Collecting farm-level agricultural data is complicated by the fact that agriculture is complex, often involving multiple products (crops and livestock), numerous plots, and a range of inputs. The fact that agriculture is a self-employment activity also means that it is difficult to ascertain income from the activity without carefully determining revenues and costs. The farm household often consumes much of the outcome of production, making the valuation of the output challenging. Furthermore, farmers are rarely involved solely in agriculture and understanding the impact of a project on farmers frequently requires looking at an agricultural household’s total livelihood strategy to see if labor or other resources have shifted as a result of the project. The logistics of collecting data can also be complicated by the fact farmers tend to be widely dispersed. In this section of the guideline, some suggestions are provided for collecting data for IE. Of course, in considering the data to collect, it is necessary to keep in mind the indicators previously identified during the baseline survey to assess impact and the approach that will be used for identifying impact. The data collected will be used to create variables that are used either as impact indicators or as part of the analysis. Our expectation is that any reasonable IE should be designed to have at least two rounds of data collection: a baseline collection already done and the upcoming post-intervention collection. More rounds of data collection are possible and can be quite useful, especially if short- and long-term impacts from the project are to be distinguished. One of the general rules of collecting data in multiple rounds is to try and maintain the same format and type for questions. Changes in the questionnaire can result in differences over time being due to changes in the way questions were posed rather than in changes in the underlying variable of interest. A common starting point for evaluating YIIFSWA is to use a standard questionnaire that has already been administered in the yam-growing areas and is already field-tested. The data collection for assessing the impact of farmer-targeted projects focuses primarily on obtaining information from detailed questionnaires of farmers, including the treated and control groups. Other information can also be collected from community-level or market-level surveys as appropriate. Here, the focus is on the data collection via questionnaires administered to farm or agricultural households, although other complementary surveys should be considered. Other considerations should focus on the following. Timing and periodicity Timing and periodicity are two of the main aspects to be considered when the procedure for data collection is being designed. The timing refers to the period (month, year) in which the data will be collected. Administering the questionnaire at the end of the season reduces the recall period and the measurement error as it enhances accuracy on the estimates of inputs used, production sold, 22 prices, and so on. The survey should be administered after the harvest for the main agricultural season has taken place. It might be problematic to collect data when the main season crop has not yet been harvested. The follow-up surveys should always be collected at the same period that the baseline was collected. Furthermore, it is important to maintain consistency checks to be sure questions refer to the same plot and crop. A benefit in addition to a reduction in recall error and thus better data is that each visit should be shorter, taking less of the farmer’s or the farming family’s time. However, this approach tends to be much more costly and there is a risk that the farmer would refuses to continue answering questions at some stage. The next aspect to consider when collecting agricultural data is periodicity. This refers to the time between the baseline administration and the follow-up surveys. The main factor that influences the periodicity of data collection is the estimated time that the project is expected to take to have an impact. This is particularly important when there is a limited budget that includes resources for only a baseline and one follow-up survey. If this is the situation, it is crucial to time the follow-up survey after the project’s impact has been expected to occur. Otherwise, the evaluation might not be able to detect any impact and would disregard the importance of the project when the actual problem was the timing of the follow-up survey. This requires a broad knowledge of the project and its effects as well as of previous empirical evidence. For instance, for technology transfers, it is expected that farmers need to learn how to incorporate the technology and how to use it appropriately; this might take time for the expected results on productivity to be produced. Pilot-testing and survey preparation Besides administering a baseline and an end-line survey, the data collection strategy must also include planning for the survey to be pilot-tested. The main purpose is to check the validity of the questionnaire by finding questions or words that might be misinterpreted, misread, or misunderstood as well as to check the functionality of the questionnaire in the field. Questionnaires often have procedures for quality assurance such as checklists that verify all questions are asked and that questions are consistent across sections. In addition, data quality assurance protocols can also be assessed. For this localized survey in which there is limited variation in the types of households, about 25-30 tests of the survey will be done. The pilot tests are also useful as a beginning to considering the logistics for survey administration. The pilot tests resemble the data collection process in the field so they can be used to consider the best manner in which to organize both data collection and data entry. 23 Survey Design, Planning, and Execution 4 This section provides the details of the end-line survey design in terms of collection methods, questionnaire design, and applied statistical analysis. This study is designed after the baseline study. The survey is necessary for calculating the impact estimators and is designed to be comparable to the baseline survey as much as possible, thereby encompassing the same survey design and instruments. Study area Following the baseline survey, the end-line survey will be done within the major yam-producing zones. The survey design will be based on a multistage, random sampling procedure, drawing on the total households from yam-growing areas of Nigeria and Ghana. Sample size determination The need for quantitative and qualitative information about households requires a statistically plausible sample of the target population. Accurate sampling is important to minimize the risk of sampling bias and to allow inferences about the population to be drawn with a level of confidence that can be statistically estimated. The Confidence Interval Approach used previously for the baseline survey will be used to estimate the sample size (Mignouna et al. 2014). Under simple random sampling, at the 95% confident level desired, the sample size n must satisfy the formula: Z 2 0.95 NP(1− P) Z 2 n ≥ ⇒ n ≥ 0.95 P(1− P) , if N>10,000 (N −1)e2 + Z 2 0.95 P(1− P) e2 Where: Z = value of the standard variate at a given confidence level and to be worked out from the table showing the area under normal curve, at 1.96 corresponding to 95% confidence level; N= Total population n ≥ 380 Provided that response rate is 100% nsrs=380/r = 380/0.95 = 400 given 95% response rate. 24 Under cluster sampling, for the results to be useably reliable, we apply a default value of design effect1 of 2.0 in Nigeria and 1.5 in Ghana as follows: ncls = δ ×nsrs Where: ncls = Sample size under cluster sampling; δ = Design effect, given the default effect δ =2.0 for Nigeria and 1.5 for Ghana (UN Stat. Division 2005); ncls (Nigeria) = 2.0× 400 = 800 ncls (Ghana) = 1.5× 400 = 600 Therefore the end-line survey will target the same total of 1400 sample households consisting of participating and non-participating households and will be conducted in the second semester 2015 in Nigeria and Ghana. However, a security challenge in Nasarawa State compelled us to select the Federal Capital Territory (FCT) as an addition. The selection of suitable comparator LGAs was undertaken to replace a few. The choice of such LGAs in FCT was predicated on their biophysical and socio-economic conditions as well as their population-related characteristics that were similar to those in Kogi, Nasarawa, and Niger States (Kasim et al. 2014). Prepared survey questionnaires (Annexes 3, 4 and 5) will be administered by trained enumerators through personal interviews and field measurements. The surveys will be conducted in the same YIIFSWA project areas as they were for the baseline. For field measurement, one out of the retained households will be randomly selected from each selected community (Annexes 1 and 2). Following the baseline, fields to be measured should amount to 200 in Nigeria and 100 in Ghana (Mignouna et al. 2014). Data collection instruments The YIIFSWA project will integrate both qualitative and quantitative methods to collect and analyze data. Quantitative results can be capable of being generalized and qualitative data will supplement quantitative IEs in providing complementary perspectives on the project’s performance in generating information that may help in understanding the mechanisms through which the project supports beneficiaries. 1 A design effect represents the combined effect of a number of components such as stratification, clustering, unequal selection probabilities, and weighting adjustments for non-response and non-coverage. A specific design effect has been applied for Nigeria and Ghana due to the different form of complex sample design employed. 25 Data will be collected for both countries by means of existing information (studies, reports, etc.,) structured questionnaires, and a set of qualitative approaches including focus groups and interviews with selected beneficiaries and other key informants. The household questionnaire (Annex 3) includes sections on (i) interview background; (ii) household composition; (iii) household identification; (iv) social capital and networking; (v) household assets; (vi) knowledge and adoption of improved/new yam varieties; (vii) crop production for all crops grown by the household during last cropping season; (viii) biotic and abiotic stress incidence; (ix) transfers and other sources of income last year; (x) household expenditure; and (xi) access to capital and support services. Community/village information will be provided using a profile form (Annex 4) in all the selected villages. The survey will capture details on existing infrastructures and facilities; active community- based groups, local decision-making systems; new varieties of yam, major livelihood strategies and constraints, and gender issues. The surveys will be facilitated by extension officers. In addition, monitoring data already collected would also be used as they constitute an important resource in the project’s IE. They will be of help in verifying which participants received the project, how fast the project is expanding, how resources are being spent, and generally whether activities are being implemented as planned. Indicators for assessing project impact Agricultural projects such as YIIFSWA are designed to improve production or the returns to agriculture. Therefore the IEs of such projects focus on production-based indicators: gross margins, crop prices, yields, productivity, agricultural investment, spending on agricultural inputs, technology adoption, changes in patterns of land use, crop, and varietal diversification, and food for home consumption. Collecting information of this type can be challenging, beginning with the definition of the sample unit: in fact, while production is often linked to multiple plots and crops, the decision- making process takes place at the household level. Although the full logic of an agricultural project should be considered, certain indicators can be more readily attributed to a given project and an IE focuses on these results. Projects may also contribute to achieve some results with a wider scope, such as a reduction in poverty rates, which may be very difficult to attribute to the project. Additionally, different indicators need to be measured and estimated at distinct time intervals. For instance, the adoption of new practices is often a short-run measure but a change in productivity is a medium to long-run measure. In considering indicators, the timing of measurement and the possibility of being able to attribute the effects to the project should be considered. The IEs often focus on examining a series of indicators to obtain a picture of the average effects of the intervention as well as the mechanism by which these effects were obtained. In analyzing agricultural production, the relationship between inputs and outputs or profitability is often examined through production or profit functions. Presumably, agricultural projects have an impact not only on production inputs and outputs but also on how they are used and combined. This is being considered in the evaluation of the YIIFSWA project. The evaluation aims to synthesize quantitative estimates of the effectiveness of AYMT demonstration plots relating to intermediate outcomes such as knowledge acquisition, adoption and diffusion of technology, and final outcomes such as agricultural yields, household income, and poverty status as depicted in the different indicators (Table 2). 26 Table 2. Indicators for Assessing Project’s Impact on Targeted Farmers. Impact Indicators Measure Agricultural income ($) Food security Agricultural Profits ($/ha) Gross Margins ($/ha) Output (tonne) Yields (Output/ha) Mechanisms of impact Price of output ($/unit) Value of harvest ($) Value or percentage of harvest lost ($ or percentage) Value or percentage of harvest for home consumption ($ or percentage) Value or percentage of harvest sold (on farm, local market, exports) ($ or percentage) Volume traded (Share of total) Market participation (Yes=1, No=0) Transaction costs ($/unit of time) Input costs ($/ha) Costs of key inputs (seeds, fertilizer, etc.) ($/ha) Family labor used (days/ha or value of days/ha) Cost of paid labor ($/ha) Costs of rented machinery (tractor, sprayer, etc.) ($/ha) Costs of rental land ($/ha) Adoption of key technology (seeds, practice, etc.) (Yes=1, No=0) Attitude and behavioral change Capacity building and development (#trainings and trainers, #MSc & PhD) Networking & sharing (#Collaborating institutions and organizations) Training of enumerators and supervisors Obtaining high quality data will be the stated aim of the survey and, as recommended by Puetz (1993), this will depend on enumerators who will be motivated, well trained, and well supervised. The structured questionnaires will be administered by enumerators under supervisors, all trained in two different methodology workshops which will be organized by IITA. The training of enumerators will be conducted for two full days and the training agenda will include project background, survey objectives, and a review of questionnaires, practice sessions, demonstrations, and logistics/ scheduling. A number of simulation sessions will be done to familiarize enumerators with questions in the household questionnaire for information to be successfully collected. Also a complete review of the questionnaires will be made on the same day in the vicinity of the sample households to permit revisits for errors to be corrected where necessary. The enumerators for each State/District will be identified after the training and testing for the whole survey. The process will be guided by factors such as (i) academic qualifications and minimum level of experience in data collection. (ii) willingness to work for long periods of time, (iii) ability both to speak the local language fluently in each given area as well as to interact with people of different ethnic groups in different environments, and (iv) familiarity with the places where the field work would be conducted. 27 Supervisors will be chosen based on extensive experience in data collection and familiarity with the survey areas. They will be trained and confirmed after an interview to make a follow-up of the whole data collection process. They will be associated with the whole process and will undertake the second quality check right in the field before the questionnaires will be accepted. Pre-testing questionnaires and guidelines will be organized and require each enumerator to complete two household questionnaires. Based on their experience, a feedback session on technique and methods will be facilitated the following day. The questionnaires and guidelines will be subsequently modified, based on enumerators’ feedback. Field data collection, data entry, and database management A field data collection schedule will be developed with the assistance of agents from the Agricultural Development Projects (ADPs) or the Ministry of Food and Agriculture (MoFA) to organize teams and assign villages according to geographic position. Geographic position in this case refers to the relative distance between the selected villages and a logical sequence for travelling without retracing routes, rather than simply those villages that were most conveniently close to the road. Because of the number of communities and the distances between them, up to three supervised teams will often be deployed in separate vehicles to each targeted administrative State/District to complete interviews. After a preliminary tour of one week, organized in surveyed areas to set up the recruitment process for potential enumerators, data collection will be undertaken from the end of 2015 in Nigeria and Ghana. IITA will be responsible for quality control of the primary data on a daily basis. Every evening, the enumerators and the field supervisors will check each household questionnaire for inconsistencies and errors. Data will be regularly packed up after a thorough check and sent for centralized data entry at IITA-Ibadan where consultant data entry clerks will be enrolled for the task. Field measurements Accurately estimating crop yields is never easy (Fermont and Benson 2011) but will improve agricultural statistics (de Groote and Traoré 2005). The study will aim at measuring crop yield directly from farmers’ fields to get more accurate estimates. The head of the household and spouse, where applicable, will be interviewed at the household level in their home for information such as characteristics of the household, available resources, yam production objectives, etc. At the field level, a structured questionnaire (Annex 5) will be employed. The field owner will respond to the oral interview for information such as production methods, varieties grown, costs of production, plans for sale and for home consumption of yam to be harvested, etc. The field-level interviews will be conducted in the selected fields. Yam yield and field area will be measured with guidance from the owner of the field. Field area measurement will be done with a Global Positioning System (GPS) receiver. Yield measurement will be based on three selected sample plots (two at the extremes and one at the center of the field) using one of the diagonal lines passing through the entire length of the given field; numbers of stands will be counted and tubers will be weighed. The yam will be purchased from the farmer at the market rate. Measurement will be done regardless of variety and fields that are‘milked’ for seed production will be skipped in yield measurement. 28 Local farmers will be used as labor for harvesting; they and the survey farmers will be paid the wage rate obtained in the community. Enumerators who will conduct the interviews and take the field area and yield measurements will be experienced scientists from IITA and the national R&D institutions in the survey countries. Field data management The questionnaires to be used on the fields will be reviewed by the YIIFSWA scientists who will lead in the field data collection. The data will be transcribed by data entry clerks who will be university graduates. After the transcription, the YIIFSWA scientists will go through the data in a verification exercise before the beginning of analyses. Verification will be a continuous process. 29 References Abubakar, G., Bakang, J.E.A and Osei, C.K. 2015. Determinants of yam minisett technology in Ghana: a case study of farmers in the Kintampo North district of Ghana. Journal of Agricultural Economics, Extension and Rural Development 3(7): 293-302. Adesina, A.A. and Zinnah, M.M. 1993. Technology characteristics, farmers’ perception and adoption decisions: a Tobilmodel application in Sierra Leone. Agricultural Economics 9: 297-311. Aighewi, B.A., Akoroda, M.O. and Asiedu, R. 2002. Seed yam production from pre-sprouted minisetts with varied thickness of storage parenchyma. African Journal of Root and Tuber Crops 5(2): 21-24. Akpoko, J.G. 2007. Analysis of Factors Influencing Adoption of Intermediate Farm Tools and Equipment among Farmers in the Semi-Arid Zone of Nigeria. Journal of Applied Sciences 7: 796-802. Amemiya, T. 1973. Regression analysis when the dependent variable is truncated normal, Econometrica 41: 997-1016. Ampofo, J.K.O., Kumar, P.L. and Seal, S.E. 2010. Integrated crop management for sustainable yam production. Pages 46–80 in Yam Research for Development in West Africa–Working Papers. IITA‐ BMGF Consultation Documents, IITA, Ibadan, Nigeria. Anderson, J.R. 2007. Agricultural Advisory Services, Background Paper for the World Development Report 2008.http://siteresources.worldbank.org/INTWDR2008/ Resources/2795087-1191427986785/Anderson_AdvisoryServices.pdf Anderson, J.R. and Feder, G. 2003. Rural Extension Services, World Bank Policy Research Working Paper 2976, World Bank, Washington DC, USA. http://econ.worldbank.org/files/24374_wps2976.pdf Anderson, J.R., Feder, G. and Ganguly, S. 2006. The Rise and Fall of Training and Visit Extension: An Asian Mini-drama with an African Epilogue. World Bank Policy Research Working Paper 3928, World Bank, Washington DC, USA. http://ideas.repec.org/p/wbk/wbrwps/3928.html Angelucci, M. and De Giorgi, G. 2009. Indirect Effects of an Aid Program: How do Cash Injections Affect Ineligibles’ Consumption? American Economic Review 99(1): 486-508. Angelucci, M. and Di Maro, V. 2010. Project Evaluation and Spillover Effects. Impact- Evaluation Guidelines, Strategy Development Division, Technical Notes No. IDB-TN-136. Inter-American Development Bank, Washington DC, USA. Baker, J.L. 2000. Evaluating the impact of development projects on poverty: A Handbook for Practitioners. World Bank Publication Abstract. New York, USA. 240 pp. Banerjee, A. 2007. Making aid work, The MIT Press, Cambridge, Massachusetts, USA. Bantilan, M.C.S, Anupama, K.V. and Joshi, P.K. 2005. Assessing Economic and Environmental Impacts of NRM technologies: An empirical application using the economic surplus approach. In Shiferaw, B., Freeman, H.A. and Swinton, S.M. (Eds), Natural Resource Management in Agriculture: Methods of Assessing Impacts. CAB International Publishing, Wallingford, UK, 245–268. Birner, R., Davis, K., Pender, J., Nkonya, E., Anandajayasekeram, P., Ekboir, J., Mbabu, A., Spielman, D., Horna, D., Benin, S. and Cohen, M. 2006. From “Best Practice” to “Best Fit”: A Framework for Analyzing Pluralistic Agricultural Advisory Services. Worldwide, DSGD Discussion Paper No. 37, IFPRI, Washington DC, USA. http://www.ifpri.org/DIVS/DSGD/dp/dsgdp37.asp. Published version: Journal of Agricultural Education and Extension 15 (4): 341-355, December 2009. 30 Bolarinwa, K.K. and Oladeji, J.O. 2009. Adoption and relevance of yam minisett technology to farmers’ indigenous practices in rain forest and derived savannah zones of Nigeria. Journal of Applied Sciences Research 5(12): 2461-2465. CIMMYT 1993. The Adoption of Agricultural Technology: A guide for survey design. Mexico, D.F.: CIMMYT Economics Program. 250 pp. de Groote, H. and Traore, O. 2005. The Cost of Accuracy in Crop Area Estimation. Agricultural Systems 84: 21-38. DiTella, R. and Schargrodsky, E. 2005. Do Police Reduce Crime? Estimates Using the Allocation of Police Forces after a Terrorist Attack. American Economic Review 94 (1): 115-133. Doss, C.R. and Morris M.L. 2001. How does gender affect the adoption of agricultural innovations? The case of improved maize technology in Ghana. Agricultural Economics 25: 27-39. Duflo, E. 2001. Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual Policy Experiment. American Economic Review 91 (4): 795-813. Duflo, E. and Kremer, M. 2005. Use of Randomization in the Evaluation of Development Effectiveness. In: G. Pitman, O. Feinstein and G. Ingram, editors. Evaluating Development Effectiveness. New Brunswick: Transaction Publishers. Feder, G., Just, R.E. and Zilberman, D. 1985. Adoption of agricultural innovations in developing countries: A survey, Economic Development and Cultural Change 33: 255-297. Feder, G., Murgai, R. and Quizon, B. 2004. Sending farmers back to school: The impact of farmer field schools in Indonesia. Review of Agricultural Economics 26 (1): 45-62. Fermont, A. and Benson, T. 2011. Estimating Yield of Food Crops Grown by Smallholder Farmers. A Review in the Uganda Context, IFPRI Discussion Paper 01097.Development Strategy and Governance Division, Washington DC, USA. Foster, J.E., Greer, J. and Thorbecke, E. 1984. A Class of Decomposable Poverty Measures. . Econometric Society, Econometrica 52(3): 761-776. Greene, W.H. 2003. Analysis. New Jersey, Prentice Hall. 778 pp. Griliches, Z. 1957. Hybrid corn: an exploration in the economics of technical change. Econometrica 25(4): 501-522. Gujarati, D.N. 2004. Basic Econometrics. Fourth edition, Tata McGraw-Hill Publishing Company Ltd, New Delhi, India. pp. 580-625. Hagmann, J., Chuma, E. Murwira, K., and Connolly, M. 1999. Putting process into practice: operationalising participatory extension. Network-Paper-Agricultural-Research-and-Extension-Network,-ODI (UK), no. 94, 23 pp. Harold, A. 2007. Improving Nutrition through Community Growth Promotion: Longitudinal Study of the Nutrition and Early Child Development Program in Uganda. World Development 35(8): 1376-1389. Heckman, J. and Robb, R. 1985. Alternative Methods for Evaluating the Impact of Interventions. Journal of Econometrics 30: 239-267. Elsevier Science Publishers B.V. (North-Holland) Holloway, G., Nicholson, C., Delgado, C., Staal, S. and Ehui, S. 2004. A revisited Tobit procedure for mitigating bias in the presence of non-zero censoring with an application to milk-market participation in the Ethiopian highlands. Agricultural Economics 31: 97-106. IITA. 1985. Root and Tuber Improvement Program. Agriculture Research Highlights 1981-1984, IITA, Ibadan, Nigeria. 31 Ironkwe, A.G. 2005. Adoption of yam minisett technology by women farmers in Abia State, Nigeria. MSc thesis in the Department of Rural Sociology and extension, Michael Okpara University of Agriculture, Umudike, Abia State, Nigeria. pp. 75-81. Ironkwe, A.G., Asiedu, R. and Unamma, R.P.A. 2007. Adoption of yam minisett technology by women farmers in Abia State. Journal of Agriculture and Social Research 7 (2). Kasim, I., Shamsudin, M.N., Yacob, R. and Radam, A.B. 2014. Technical Efficiency in Maize Production and its Determinants: A Survey of Farms across Agro-ecological Zones in Northern Nigeria. Trends in Agricultural Economics 7: 57-68. Maredia, M.K. 2009. Improving the Proof: Evolution of and Emerging Trends in Impact Assessment Methods and Approaches in Agricultural Development, IFPRI Discussion Paper 00929. 2020 Vision Initiative.http://www.ifpri.org/millionsfed Maredia, M.K., Byerlee, D. and Anderson, J. 2000. Ex Post Evaluation of Economic Impacts of Agricultural Research Programs: A Tour of Good Practice. Paper presented to the Workshop on “The Future of Impact Assessment in CGIAR: Needs, Constraints, and Options”, Standing Panel on Impact Assessment (SPIA) of the Technical Advisory Committee, Rome, 3-5, May 2000. Maroya, N., Asiedu, R., Kumar, P.L., Lopez-Montes, A., Orchard, J. and Ndiame, F. 2014. YIIFSWA Project Description.YIIFSWA Working paper Series No. 1. Yam Improvement for Income and Food Security in West Africa, International Institute of Tropical Agriculture, Ibadan, Nigeria. 18 pp. McDonald, J.F. and Moffit, R.A. 1980. The use of Tobit analysis, Rev. Econ.Stat. 61: 318-321. Mignouna, D.B, Abdoulaye, T., Alene, A., Akinola, A.A., and Maroya, N. 2014. Baseline protocols. The case of Yam Improvement for Income and Food Security in West Africa (YIIFSWA) Project.YIIFSWA Working Paper series No. 4. Yam Improvement for Income and Food Security in West Africa, International Institute of Tropical Agriculture, Ibadan, Nigeria. 80 pp. Mutandwa, E., Gadzirayi, C.T. and Zingwe, F. 2007. Factors affecting Adoption of Long Staple Cotton Variety among Smallholder Farmers in Zimbabwe. Eastern Africa Social Science Research Review 23(1): 1-14. Nkamleu, G.B. 2004. Productivity growth, technical progress, and efficiency change in African agriculture. Africa Development Review 16 (1): 203-222. Nkamleu, G.B. and Tsafack, R. 2007. On measuring indebtedness of African countries. African Finance Journal 9 (1): 21-38. OECD-DAC 2002. Glossary of Key Terms in Evaluation and Results based Management. Paris: OECD. http://www.oecd.org/dataoecd/29/21/2754804.pdf. Oladele, I. 2005. A Tobit Analysis of Propensity to Discontinue Adoption of Agricultural Technology among Farmers in Southern-western Nigeria. Journal of Central European Agriculture 6(3): 246-254. Orkwor, G.C., Asiedu, R. and Ekanayake, I.J. 2000. Food Yams: Advances in Research, IITA, Ibadan and NRCRI, Umudike, Nigeria. 249 pp. Paz, R., Dorward, A., and Douthwaite, B. 2006. Methodological guide for evaluation of pro-poor impact of small-scale agricultural projects. Centre for Development and Poverty Reduction. Imperial College, London, UK. Polson, R.A. and Spencer, D.S.C. 1991. The technology adoption process in subsistence agriculture: the case of cassava in southwestern Nigeria. Agricultural Systems 36: 65-78. 32 Puetz, D. 1993. Improving Data Quality in Household Surveys, in: Data Needs for Food Policy in Developing Countries: New Directions for Household Surveys, edited by J. von Braun and D. Puetz. International Food Policy Research Institute, Washington DC, USA. Ransom, J.K., Paudyal, K. and Adhikari, K. 2003. Adoption of improved maize varieties in the hills of Nepal. Agricultural Economics 29: 299-305. Ravallion, M. 2009, Should the Randomistas Rule? Economists Voice 6(2): Article 6. Available: www. bepress.com/ev/vol6/iss2/art6. Shakya, P.B. and Flinn, J.C. 1985. Adoption of modern varieties and fertilizer use on rice in the Eastern Terai of Nepal. Journal of Agricultural Economics 36: 409-419. Strauss, D., Bednar, L. and Mees, R. 1989. Do one percent of forest fires cause ninety-nine percent of the damage? Forest Science 35: 319-328. Suresh, D.M., McKenzie, D. and Woodruff, C. 2007. Returns to Capital in Microenterprises: Evidence from a Field Experiment. World Bank Policy Research Working Paper 4230, Washington DC, USA. Tobin, I. 1958. Estimation of relationship for limited dependent variables. Econometrica 26: 24-36. Todd, P.E. 2008. Evaluating Social Programs with Endogenous Program Placement and Selection of the Treated. In Schultz, T.P. and Strauss, J.A. (Eds.), Handbook of Development Economics (Vol. 4: 3848-3894). Elsevier Science Publishers B.V. (North-Holland). UNEG (United Nations Evaluation Group). 2013. Impact Evaluation in UN Agency Evaluation Systems: Guidance on Selection, Planning and Management. Guidance Document. United Nations. New York, USA. 43 pp. United Nations Statistics Division. 2005. Designing Household Survey Samples: Practical Guidelines, ST/ESA/SER.F/98,United Nations. New York, USA. Van den Berg, H. and J. Jiggins. 2007. Investing in Farmers–The Impacts of Farmer Field Schools in Relation to Integrated Pest Management. World Development 35 (4): 663-686. Wiredu, A.N., Gyasi, K.O., Buah, S.S.J., Asante, B.O. and Mensah-Bonsu, A. 2012. Factors affecting proportions of land allocated to the minisett technology by yam producers in Northern Ghana. African Journal of Agricultural Research 7(29): 4158-4166. Wooldridge, J. 2001. Econometric Analysis of Cross Section and Panel data, The MIT Press, Cambridge, Massachusetts, USA. 33 Annexes 1. YIIFSWA Project Communities in Nigeria. States LGAs Communities BENUE Agatu Enumgba BENUE Igoje BENUE Obagaji BENUE Okokolo BENUE Oshigbudu BENUE Gboko Adzer-Nor BENUE Akpager BENUE Luga BENUE Tchowanye BENUE Yandev BENUE Katsina-Ala Abaji BENUE Gbor BENUE Ikowe BENUE Sai BENUE Tor-Donga BENUE Otukpo Adoka BENUE Ogali BENUE Otobi BENUE OtukpoNobi BENUE Uwaba-Aokwu BENUE Tarka Agudu BENUE Gwarche BENUE Nyambee BENUE Tyiotyu BENUE Wannune BENUE Ukum Ayati BENUE Chito BENUE Kyado BENUE Vaase BENUE Zaki-Biam EBONYI Ezza North Ekka EBONYI Inyere EBONYI Nkomoro EBONYI Ogboji EBONYI Umuoghara EBONYI Ivo Akaeze-Ukwu EBONYI Ihenta EBONYI Iyuoji EBONYI Mgbede EBONYI Umobor EBONYI Izzi Agbaja EBONYI Agbanyim EBONYI Igbeagu EBONYI Ndieze EBONYI Yimaegu EDO Esan Illushi EDO Ivue EDO Obeidu EDO Onogholo EDO Oria 34 Annex 1. YIIFSWA Project Communities in Nigeria contd. States LGAs Communities EDO Orthioromwon Iguemokhua EDO Owuo EDO Ugoniyekonhonma EDO Umoghun-Nokhwa EDO Uromehe EDO Owan East Arokho EDO Ihiebe EDO Irbiaro EDO Ohanmi EDO Warake ENUGU Aninri Mpu ENUGU Ndiaboh ENUGU Nenwe ENUGU Oduma ENUGU Opanku ENUGU Awgu Agbogugu ENUGU Agwu ENUGU Amoli ENUGU Ifite ENUGU Maku ENUGU Enugu East Alulu ENUGU Amorji ENUGU Ibagwa ENUGU Nkwugbo ENUGU Ugwogo ENUGU Igbo-Eze Aguibeje ENUGU Amube ENUGU Okpo ENUGU Onicha ENUGU Umuopu ENUGU Igbo Etiti Ekwegbe ENUGU Ohodo ENUGU Ozalla ENUGU Ukehe ENUGU Umunko ENUGU Udenu Imilike ENUGU Obollo Eke ENUGU ObolloEtiti ENUGU Ozalla-Ezimo ENUGU Umundu ENUGU Uzo-Uwani Abbi ENUGU Nimbo ENUGU Nrobo ENUGU Opanda ENUGU Uvuru FCT Abaji Agyana FCT Makana FCT Nuku FCT Pandagi FCT Yewuni FCT Gwagwalada Ibura II FCT Luda 35 Annex 1. YIIFSWA Project Communities in Nigeria contd. States LGAs Communities FCT Pagadan FCT RaphinZuti FCT Wura FCT Kuje Chibiri FCT Kiyi FCT Shadarbi FCT Shazi FCT Tarkarba FCT Kwali Ashara FCT Kilankwa I FCT UboSaidu FCT Ubosharu FCT Yambabu KOGI Ibaji Odogwu KOGI Ogaine KOGI Ojuba KOGI Onyedega KOGI Ujeh KOGI Idah Ajibaja KOGI Ekwokata KOGI Ichala KOGI Ijobe KOGI Ojigagala KOGI Omala Abejukolo KOGI Ajiyolo KOGI Bagaji KOGI Icheke? KOGI Odoh KOGI Yagba East Ejuku KOGI Imela KOGI Jege KOGI Ponyan KOGI Takete-Isao NASARAWA Lafia Adogi NASARAWA Agudu NASARAWA Assakio NASARAWA BukanBuzu NASARAWA Bukan Koto NASARAWA Nasarawa Gadabuke NASARAWA Karmu NASARAWA Kwoho NASARAWA Laminga NASARAWA MararabaUdege NASARAWA Obi Agyaragu NASARAWA Daddere NASARAWA Kpangwa NASARAWA Obi NASARAWA Zherugba NIGER Bosso Beji NIGER Garatu NIGER Garusu NIGER Gbaiko NIGER Kampala 36 Annex 1. YIIFSWA Project Communities in Nigeria contd. States LGAs Communities NIGER Gurara Bonu NIGER Diko NIGER Lambata NIGER Lefu NIGER Tufa NIGER Lapai BirninMaza NIGER Gabi NIGER Gulu NIGER Gupa NIGER Lapai NIGER Mashegu BabbanRamin NIGER Makari NIGER Mashegu NIGER Masuchi NIGER Sahorami NIGER Rafi Karaya NIGER Katako NIGER Madaka NIGER Sambuga NIGER Tegina NIGER Shiroro Gwada NIGER Kadna NIGER Pina NIGER She NIGER Zumba NIGER Tafa Azhi NIGER Garam NIGER Gyedna NIGER Ijagwari NIGER SabonWuse OYO Irepo Adagbangba OYO Gudu OYO Nufe OYO Sooro OYO Welewele OYO Olorunsogo Alawa OYO Bi-Alaso OYO Dogo OYO Igbeti OYO TesiGarubar OYO Orelope Bonni OYO Igbope OYO Kajola OYO Oloko OYO Sooro 37 2. YIIFSWA Project Communities in Ghana. Regions Districts Communities ASHANTI Ejura-Sekyedumase Bisiw 1 ASHANTI Bisiw2 ASHANTI Bompa ASHANTI Ejura Nkwanta ASHANTI Hiawoanwu ASHANTI Kasei ASHANTI Kramokrum ASHANTI Krampong ASHANTI Kropong ASHANTI Leafu Kura ASHANTI Mesuo ASHANTI Nkrama ASHANTI Nokreasa ASHANTI Nyinasei ASHANTI SamariNkwanta ASHANTI Sunkwae BRONG-AHAFO Atebubu-Amantin Akyeremade BRONG-AHAFO Amanfrom BRONG-AHAFO Asanteboa BRONG-AHAFO Badukrom BRONG-AHAFO Boniafo BRONG-AHAFO Densi BRONG-AHAFO Duabone 1 BRONG-AHAFO Duabone 2 BRONG-AHAFO Kafaano BRONG-AHAFO Kumkumso BRONG-AHAFO Lailai BRONG-AHAFO Mem BRONG-AHAFO Morochusu BRONG-AHAFO Nwowam BRONG-AHAFO Old kronkrompe BRONG-AHAFO Patuda BRONG-AHAFO Praprabon BRONG-AHAFO Primukyea BRONG-AHAFO Sampa BRONG-AHAFO Tintare BRONG-AHAFO Watro BRONG-AHAFO Kintampo Aduma BRONG-AHAFO Alassankura BRONG-AHAFO Asantekwa BRONG-AHAFO Asuma Kura BRONG-AHAFO Attakura BRONG-AHAFO Bablioduo-Kokomba BRONG-AHAFO Badu Krom (Kofi) BRONG-AHAFO Basabasa BRONG-AHAFO Ben Krum BRONG-AHAFO Busuama BRONG-AHAFO Chiranda BRONG-AHAFO Dawadawa BRONG-AHAFO Gulumpe BRONG-AHAFO Kadelso BRONG-AHAFO kaka BRONG-AHAFO Kandige BRONG-AHAFO Kawampe BRONG-AHAFO KurawuraAkura BRONG-AHAFO Mansira BRONG-AHAFO Miawani BRONG-AHAFO NanteZongo BRONG-AHAFO Nyamebekyere 1 BRONG-AHAFO Nyamebekyere 2 BRONG-AHAFO Sogliboi BRONG-AHAFO Suronuasi BRONG-AHAFO Taidifufuo BRONG-AHAFO Techira 1 BRONG-AHAFO Techira 2 BRONG-AHAFO Yaara BRONG-AHAFO Yabraso 38 Annex 2. YIIFSWA Project Communities in Ghana contd. Regions Districts Communities NORTHERN East Gonja Abrumase NORTHERN Adamupe NORTHERN Bau NORTHERN Bunjai NORTHERN Dagbabia NORTHERN GrunshieZongo NORTHERN Jemitutu NORTHERN Kakoshi NORTHERN Kalande NORTHERN Mbawupe NORTHERN Katanga NORTHERN Kigbatito NORTHERN Kijewu NORTHERN Kitoe NORTHERN Kpolo NORTHERN Kumburupe NORTHERN Latinkpa NORTHERN Masaka NORTHERN Mbawudo NORTHERN Nakpaye NORTHERN shishipe NORTHERN Talkpa NORTHERN Tunga NORTHERN Mion Gunsi NORTHERN Kulunkpegu NORTHERN Mahakpi NORTHERN Mbatinga NORTHERN Ndiyuriyili NORTHERN Puriya NORTHERN Salankpang NORTHERN Sang NORTHERN Sanze NORTHERN Zakpalsi 39 3. Household level questionnaire International Institute of Tropical Agriculture (IITA) Yam Improvement for Income and Food Security in West Africa (YIIFSWA) TECHNOLOGY ADOPTION & IMPACT SURVEY HOUSEHOLD QUESTIONNAIRE Nigeria and Ghana Part A. Interview Background 1. Field supervisor’s name ____________________________ 2. Date Checked ___________________ 3. Data Entry clerk’s name __________________________ 4. Date Entered ___________________ Respondent’s Telephone Number: _____________________________________ 5. Enumerator’s name: _________________________________ 6. Date of interview: __________________ 7. Country: _____________________ 8. State/Region: ____________________________ 9. LGA/District: _________________________ 10. Village/Community: ______________________________ GPS readings at the house of respondent Latitude (N/S) Longitude (W/E) Altitude in meters . 0 . 0 [The respondent must be the head or de-facto head of the household] 11. Name of respondent: ______________________________________________________________________ 12. Gender of respondent: ______ [1] Male [0] Female 13. Age of respondent (in years): ______________ 14. Education (in years with 0= None/Illiterate): _____________ 15. Religion of the household head: _____ (1. No religion/atheist/traditionalist; 2.Christian; 3.Muslim; 4. Other) 16. Experienceof the household head in growing yam (years): _____________ 17. Is the respondent the head of the household? ____ [1] Yes [0] No 18. Total number of people in the household: ______________, Females out of the total: ___________________ 19. Have you benefited directly or indirectly from YIIFSWA project? _____ (1=Direct; 2 = Indirect; 3=Both) 20. If you benefited from YIIFSWA project from Question 19, indicate how. _____ (1= Training in minisett tech- nique; 2= Training in vine cutting technologies; 3= Training in business plan development for pre-basic and basic seed producers; 4= Training in business plan development for seed producers; 5=Training in business plan development for yam producers; 6=Provided with seed tubers; 7=Provided with QDS/pre- basic/basic materials for seed production; 8=Provided with plantlets; 9=Training on high-ratio seed yam propagation techniques; 10=Production of certified seed yam; 11=Training in seed yam quality control and certification; 12=Improved yam storage facilities; 13= Others (specify: ______________________________) 21. Have you put the benefit gained into use? ____ (1= Yes; 0= No) 40 22. If Yes to question 21 above, how many people have benefited from you? _________ 23. If you benefited from YIIFSWA project, how has it helped you?______ (1= Yam production increased; 2=Area of yam produced increased; 3= Output per unit of area increased; 4= Reduced losses; 5= Improved quality and safety of processed products; 6=Application of quality management protocols system in seed yam production; 7= Others (specify: ___________________________________________________________) 24. In which year did you start benefiting from YIIFSWA project? ____ (1=2012, 2=2013; 3=2014; 4=2015) 25. How satisfied are you with the project benefit(s)? _____ (1= Very satisfied; 2= Satisfied; 3= Not satisfied) 26. If not satisfied with the project benefit(s), why? ____ (1= Lack/shortage/unavailability of input; 2=Not enough time for learning; 3= Not convinced; 4= Poor quality of planting material received; Others, specify: __________________________________________________________________________________________) 27. Type of toilet used: _______________ (1. Flush toilet private; 2. Flush toilet shared; 3. Ordinary pit latrine private; 4. Ordinary pit latrine shared; 5. No toilet/use open air) 28. Main walling material of main residential house: __________ (1. Burned bricks; 2. Unburned bricks; 3. Mud bricks; 4. Concrete blocks; 5. Pole and mud; 6. Timber; 7. Sticks and grass; 8. Iron sheets; 9. Other, specify………………………………………….............) 29. Main roofing material of main residential house: ________ (1. Grass thatch; 2. Iron sheest; 3. Tiles; 4. Asbestos; 5. Other, specify…………………….………………) 30. Taking into consideration ALL food sources (own food production + food purchases + help from different sources + food hunted from forest and lakes, etc.), how would you assess your family’s food consumption in the past 12 months? _______________ (1. Food shortage through the year, 2.Occasional food shortage, 3. No food shortage but no surplus, 4. Food surplus) 31. In case of food shortage from 30 above, what is the most important coping strategy used? _________ 1. Rely on less preferred foods; 2. Limit the variety of foods eaten; 3. Limit portion size at meal‐times; 4. Reduce number of meals eaten in a day; 5. Restrict consumption by adults for small children to eat; 6. Borrow food, or rely on help from a friend or relative; 7. Have no food of any kind in your household; 8. Go to sleep at night hungry because there is not enough food; 9. Go a whole day and night without eating anything; 10. Seek jobs inside the community; 11. Migrate to urban centers in search of non-farm jobs; 12. Other, specify: …………………………………………………………………) 32. Distance from residence to the nearest primary school in the community? _______ (minutes of walking time and NA if none) 33. Distance from residence to the nearest farm? _________ (minutes of walking time) 34. Distance from residence to the local market? _________ (minutes of walking timeand NA if none) 35. What means of transport do you use most frequently to get to the local market from your residence? ______ (1=Walking; 2=Bicycle; 3=Motorcycle; 4=Tractor; 5=Vehicle; 6=Cart; 7=Other; 8=NA) 36. Quality of road to the main market (district) ___ (1=Very poor; 2=Poor; 3=Average; 4=Good; 5=Very good) 37. Average one-way transport cost (/person) to the main market using a car __________(Naira/Cedi/pers.) 41 38. Distance to the nearest health center from residence _________ (minutes of walking time) 39. Main source of drinking water _____ (1=Piped/tap; 2=Deep well protected and covered; 3=Deep well unprotected & uncovered; 4=Stream; 5=River; 6=Dams; 7=Ponds or floods; 8=Borehole) Note: protected refers to water sources internally plastered and covered with a cap of wood, stone, or concrete) 40. Distance to main water source for drinking ________ (minutes of walking time) 41. Are you involved in any projects that are going on in the community? ___ [1] Yes [0] No 42. If Yes to question 41, what kind of projects are they? (Tick as appropriate): a. Agricultural extension services b. Microcredit c. Community Health Volunteer Training d. Water supply e. NGO (Non-governmental Organization) starting new activities f. Other projects, which? _____________________________________________________ 42 43 Part B. Household Composition 1. We are interested in knowing more about the composition of your household (all the people living in the same compound, eating from the same “pot” and working on the family farm) Name Gender Age Relation Marital Education Indicate type of Number of Number of For those under the age (Start with the house- 0=Female in to head: Status level off-farm income months liv- months (in a of 5, any illness in the last hold head) 1=Male Years (See Code (See Code (in years HH member is ing at home year) available one year? below) below) with 0= Il- earning in the last for farm work (See Codes below) literate) (Code below) 12 months 3 years ago Now 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 0=Head 0=Single 0=Petty trading 0. No disease 1=Spouse 1=Married 1=Teaching 1. Fever/malaria 2=Parent 2=Widowed 2=Masonry/carpentry 2. Dysentery/diarrhoea 3=Child/grand 3=Separated 3=Nursing 3. Respiratory problems child 4=Divorced 4=Arts and crafts 4. Measles 4=Nephew/Niece 5=Driving 5. Typhoid fever 5=Son/daughter- 6=Fitting mechanic 6. Under-nutrition in-law 7=Farm labor 7. Tuberculosis 6=Brother/Sister 8=Civil service 8. Lifetime disease/Disorder 7=other relative 9=Other 9. Other specify 10=N/A Part C. Social Capital and Networking 1. Have you and/or your spouse belonged to any farmers’ associations in your Village/Community in the last 3 years? ____ [1]=Yes [0] No 2. If Yes to Question 1, Which of the following association(s) do you belong to and what level of involvement? Association Yes=1, If yes, Role in the Major activity (Use No=0 year association codes) joined 1=Leader, 1. Input access/pro- 2=Active duce marketing; 2.Seed member, production; 3. Farming; 3=Ordinary 4.Savings and Credit; 5. member Funeral group; 6.Church group /congregation; 7.Input credit Input supply/farmer coops/union Cooperative Yam producer and marketing group Other crop/seed producer & marketing group/ coop Local administration Farmers’ Association Women’s Association Youth Association Religious association Savings & Credit group Funeral association Government team Cooperative farming Other, specify: _____________________ 3. Number of years you have been living in this community: ______________ 4. Number of yam traders that you know in this community who could buy your yam: __________________ 5. Number of yam traders that you know outside this community who could buy your yam: _______________ Part D. Gender Role Section 1. Household decision-making Per activity mentioned, please use only one option Head Spouse Jointly Children along the row alone alone (Head and spouse) Crops to plant (1=Yes) Type of yam variety to plant (1=Yes) Purchase of yam seeds to plant (1=Yes) Allocation of labor under yam (1=Yes) Amount of yam produced to be consumed (1=Yes) Amount of yam produced to be sold (1=Yes) Food security coping mechanism to use in case of food shortage (1=Yes) 44 Section 2. Household farming operations Farm Operation Who decides on the following? Who did most of the farming operations? 1-Head alone; (Please mark only one option per farming operation) 2- Spouse alone; Men mostly Women Both Children 3- Jointly; (1=Yes) mostly equally (1=Yes) 4- Children (1=Yes) (1=Yes) 5-NA for unused operation Land clearing Seedbed preparation Yam treatment Planting Fertilizer application Mulching Staking Weeding Harvesting Transport Storage Marketing Part E. Household Assets Section 1. Ownership of productive and household assets Number Estimated unit value in terms of how much Asset you would receive from the sale (Naira/Cedi) (if no equipment, put zero) (if more than one item reported in column 2, take average price) Cart Axe Machete/ cutlass Hoe Sprayer Grain mill Pump Spade or shovel Radio CD Player Television set Cell phone Stove Bicycle Motorbike Car Tractor Jewellery Wooden box Metal box Bed Chair Table Thatched house Corrugated iron sheet house Fish pond Sofa Panga knife Other, specify...................... 45 Section 2. Livestock ownership Livestock owned Type Number owned Type Number owned 1.Cattle 7.Poultry (Chickens, Guinea fowl, Ducks) 2.Donkeys 8.Doves/pigeons 3.Horses 9.Pigs 4.Goats 10.Other (Specify1: ________________________) 5.Rabbits 11. (Specify2: _______________________) 6.Sheep 12. (Specify3: _______________________) Section 3. Land holding during the last cropping year Land holding Land holding Land holding Land category Land holding (ha) share for ware share for seed share for women yam (%) yam (%) (%) 1. Own land used (A) 2. Rented in land (B) 3. Rented out land (C) 4. Borrowed in land (D) 5. Borrowed out land (E) 6. Total owned land (A+C+E) 7. Total operated land (A+B+D) Section 4. Yam storage during the last cropping year Number of months your Type of storage used Percentage lost at the end of storage yam is stored Rotting (%) Sprouting (%) Other: …………………… (Use codes below) …………….. months Codes: 1=Improved storage system, 2=Traditional room storage; 3=Under trees; 4=Raised sheds in the field; 5=Yam barns in the compound ; 6=Raised huts; 7=Left in the soil after maturity; 8= Other (specify: …………………………..) Part F. Knowledge and adoption of new yam technologies Section 1. Awareness, adoption, and disadoption of yam technologies 1. Indicate on the Table below the different yam varieties planted How many different varieties What were the main varieties of If you divide the total plots into 10 of yam did you/will you plant? yam planted last season? parts, how many parts are under 3 years Last In the First variety Second Third your main yam variety? ago season future _______ /10 Code for Ghana Code for Nigeria 1-Pona 6-Nanto 11- Chinchinto 16- 18-Mankro- 21-Amula 26-Meccakusa Matches/ npona 2-Lariboko 7-Kokropa 12-Maamak- Seidu/ 22-Ame 27-Hembakwase umba Afasie/ 19- Aso- 3-Dente 8-Nooma Blu/ bayere/ 23-Pepa 28-Danancha 13-Serwaa/ 4-Akaba 9-Akwa Afibetua Water Auntie/Ako- 24-Gwagwa 29-Water yam yam sua 5-Lelee 10-Enkanfo 14-Lobare/ 25-Yangode 30-Others:__________ Dorban 17-Mu- 20-Oth- chumudu/ ers:_______ 15-Kwaseko- hwe Araba/ Asana/ Moonye/ Moninyo 46 2. Have you ever planted any new variety of yam during the last 5 years? ___ [1] Yes [0] No 3. If No to question 2, why have you never planted any new yam variety? [1] N/A [2] Not heard of any new varieties [3] No access to planting material [4] No money to buy the planting material [5] Satisfied with the local varieties I plant [6] Simply not interested in experimenting with new varieties [7] Not seen any demonstration to show superiority of new varieties [8] Other: __________________________________________________ If No to question 2, skip to question 15, Otherwise continue with question 4 below. 4. How many years ago did you plant a new variety of yam for the first time? _________ 5. What was the name of the new variety you planted for the first time? ________________ 6. What was the source of information about the new variety? [1] Fellow farmer [2] Local retail shop [3] Ministry of Agric. Extension agent [4] Seed company staff ___________ [5] Research Institute [6] NGO (specify) _______________ [7] Radio [8] Television [9] Newspaper [10] Other (specify) ______________ 7. What was your source of planting material? [1] Do not remember [2] Free plant. mat. from a neighbor [3] Free plant. mat. from government program [4] Free plant. mat. from an NGO program [5] Purchased from a Seed company [6] Purchased from NGO [7] Purchased from Ministry of Agriculture [8] Purchased from another farmer [9] Purchased from market [10] Purchased from an agro-dealer [11] Other: ________________________________ 8. What was the reason for your choice of source of planting material? [1] No reason [2] Cheaper source [3] Available source [4] Lack of cash [5] Near homestead [6] Free source [7] Other: _________________________________________________ 9. Have you been planting new yam varieties since then (continuously)? ____ [1] Yes [0] No 10. If No to question 9, how many years ago did you discontinue planting? ________________ 47 11. If No to question 9, why did you discontinue planting? [1] N/A [2] Preferred variety no longer available [3] No cash to purchase plant. mat. [4] Not satisfied with performance of the varieties [5] Depressed prices [6] Other: ___________________________________ 12. After discontinuing, in which year did you resume planting any new yam variety? _________ 13. Which variety did you plant when you resumed planting? _______________________________ 14. Why did you resume planting new yam varieties? [1] N/A [2] New varieties satisfied my demand [3] Local varieties performing too poorly [4] Convinced by extensionist [5] Other: ______________________________________ 15. Will you purchase any new variety for planting in the future? _____ [1] Yes [0] No 16. If No to question 15, why will you not purchase yam planting material in the future? [1] N/A [2] No cash to purchase plant.mat. [3] Will not obtain preferred variety [4] No plant. mat. retailer within locality [5] Already satisfied with the plant. mat. I have [6] Other (specify):____________________ Section 2. Awareness of adaptive yam minisett technology, adoption, and disadoption Awareness and adoption related questions Adaptive yam Improved yam minisett technology storage technology (If No to the below row for both technologies, skip the table) Ever heard (1=Yes, 0=No) Year first known or heard (YYYY) Main source of information (Codes A) Ever planted/used (1=Yes, 0=No) If ever planted/used, year first planted/used (YYYY) If ever planted/used, area/volume first planted/used (ha/liter) If ever planted/used, source of sett/storage first planted/used (codes C) If ever planted/used, quantity of sett/yam first planted/stored (indi- cate number of setts or kg or baskets) If ever planted/used, main means of acquiring the setts/technology first planted/used (Codes D) If ever planted/used, yield/quality of first planted /used(indicate number of minitubers or kg or use Codes E) If ever planted/used, number of years minisetts have been planted/ months yam stored (Number) Did you plant/store minisett/yam last season (2014)? (1=Yes, 0=No) If never planted/stored, give main reason why (Codes B) 48 Codes A Codes B Codes C Codes D 1. Government extension 1. Technical knowledge not 1. On-farm trials 1. Gift/free available 2. Farmers’ Coop/group 2. Extension demo fields 2. Borrowed seeds 2. Seeds/technology not avail- 3. IITA able 3. Farmers’ groups/coops 3. Bought with cash 4. CRS, MSHR, New NGOs 3. Lack of cash/credit to acquire 4. Local seed producers 4. Payment in kind seeds/technology 5. CRI, GLDB, NRCRI, 5. Seed retailers 5. Exchange with other seeds 4.Susceptible to diseases/pests 6. Seed company/grain stockist 6. Private seed suppliers 6. Research Institute/IITA 5. Poor taste 7. Relative/ Neighbor 7.MoFA/ADP 7. CRS/MoFA/MSHR 6. Low yielding/quality 8. Radio/newspaper/TV 8.Club/association 8. Other, specify__________ 7. Price too high 9. Local input provider 9. Farmer-to-farmer seed 8. No market exchange 10. Other, specify__________ Codes E 9. Poor storability 10. Provided free by NGOs/govt 1.Lower 10. Lack of enough land 11. Other (specify)___________ 2.Unchanged 11. Requires high skills 3. Better 12.Content with current 13. Other, specify____________ 49 50 Part G. Crop Production for All Crops Grown by the Household during last Cropping Season Section 1. Characteristics, investment, and input use Definitions: A field is a piece of land physically separated from others; a plot is a subunit of a field. If more than one crop is grown on a field (that is, on different plots), repeat the code in next row and use plot code. If the plot is intercropped, use same row and separate the different intercrops by commas. Consider only 3 main intercrops if you encounter more than 3 crops on a plot and start always the first row(s) with yam. Field loca- tion name (as called by farmer) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Codes A Codes B Codes C Codes D Codes E Codes F Codes G Codes H 3. Grass strips 6.Minimum till 1. Walking 5. Other 1. Owned 4. Borrowed in 0. Women 1. Good 1. Gentle slope (flat) 1. Shallow 1. Black 4. Grey 0. None 4. Trees on 7.Soil bunds 2. Bicycle 2. Rented in 5.Borrowed out 1. Men 2. Medium 2. Medium slope 2. Medium 2. Brown 5. Other, 1. Terraces boundaries 8.Stone bunds 3. Motorcycle 3. Rented 6. Other, specify…. 2. Both equally 3. Poor 3. Steep slope 3. Deep 3. Red specify… 2. Mulching 5. No tillage 9. Box ridges 4. Vehicle/ tractor out 10.Other, specify… Serial number Field code (start with one next to resi- dence) Plot code Plot size (Ha) Intercrop? (0=No; 1=Yes) Crop rotation (0=No; 1=Yes) Crop(s) grown (Use Annex 1 codes) Crop variety (0=local; 1=new) Percent of area un- der each intercrop (e.g. 50,50) Transport used from residence to field Codes A Field distance from residence (walking minutes) Plot ownership Codes B Plot manager Codes C Soil fertility Codes D Soil slope Codes E Soil depth Codes F Soil type/colour Codes G Soil & water conser- vation method Codes H Crop residue left on plot 1=Yes; 0=No 51 Section 2. Crop harvested, utilization of crop produced, and household food security Different from Section 1 above: i.e., record one row per crop (e.g., add production from all yam fields together) and start with yam NB. Report the number of tubers for yam instead of kg From the total available stock (Column 5)… Total har- Carryover Total avail- Crop vested Seeds used Gifts, donations mount left in Form stock from able stock A for this Quantity In-kind payments during last given out dur- Consump- store before next (Use codes last season previous son sold (kg) (labor, land and oth- cropping ing last cropping tion during harvest in Annex 1) Codes A harvest sea ers) paid during last season season last cropping (kg) (kg) (kg) cropping season season (kg) (kg) (kg) (kg) (kg) 1 2 3 4 5=2+4 6 7 8 9 10 11=5-6-7-8-9-10 Codes A: 1. Fresh/green; 2. Dry 52 Section 3. Marketing of crops 1. Different from Sections 1 & 2 above: i.e., record one or more than one, according to market type transaction (e.g., add production from all crop fields together by market type) and start with yam NB. Report the number of tubers for yam instead of kg Quantity sold (kg) Crop Market (Use codes in type Month sold (sum by crop Who sold? Price Buyer Relation Mode of trans- Time taken to Annex 1) Codes C (Naira/ Codes to buyer port get to the market Codes A Codes B should be equal to Actual transport Column 6 of Sec- Cedi/kg) D Codes E Codes F (minutes) cost (Naira/Cedi) tion 2) 1 2 3 4 5 6 7 8 9 10 11 Codes A Codes B Codes C Codes D Codes E Codes F 1. Farm gate 1. January 7. July 0. Female 1. Farmers’ group/union / 6. Urban grain trader/wholesaler 1. No relation but not a long- 1. Bicycle 2. Community market 2. February 8. August 1. Male coop/club 7. Exporter time buyer 2. Hired vehicle 3. Main/district market 3. March 9. September 2. Both 2. Consumer or other farmer 8. Miller 2. No relation but a long-time 3. Public transport 4. April 10. October 3. Rural assembler 9. FRA buyer 4. Donkey 5. May 11.November 4. Brokers/middlemen 10. Other, specify………….…. 3. Relative 5. Oxen/horse cart 6. June 12. December 5. Rural grain trader/whole- 4. Friend 6. Back/head load saler 5. Money lender 7. Other, speci- 6. Other, specify…… fy……... 2. What is the maximum amount of money you would be willing to pay for a seed yam variety that has the desired qualities and is sufficient for planting one hectare? .......................... (Naira/Cedis) 3. What is the most important objective for growing yam for your household______ (1=Sale; 2=Food; 3=Both) 4. How many times did you harvest your field______ (1=once; 2=twice; 3=more than twice; 4=others, specify: ___________________________________________) 5. Where did (would) you carry the yam harvested? ___ (1=Home; 2=Market; 3=Store in the field; 4=Other, specify: ___________________________________) 6. Main type of seed yam used? ________ (1=Whole tuber; 2=Sliced tuber; 3=Milked tuber) 7. If you divide the entire yam you harvest into 10 parts, how many parts do you use as seeds? Use as seeds _______ Parts out of 10 8. If you divide the yam you sell into 10 parts, how many parts will you take to market and how many parts will traders come to you and buy? To market _______ Parts out of 10 Traders come for _______ Parts out of 10 ________________ Total 10 9. If you divide the yam you sell into 10 parts, how many parts will you sell immediately after harvest and how many parts will you store and sell later? Sell immediately at harvest _______ Parts out of 10 Store and sell later _______ Parts out of 10 ________________ Total 10 10. Do you process the yam you produce? ____ (1=Yes; 0=No) 11. If Yes to Question 10 above, If you divide the entire yam you harvest into 10 parts, how many parts will you process? To process _______ Parts out of 10 12. What is the main challenge/constraint confronting the sales of your yam products in this community and in other places? ___ (1=Transportation costs; 2=Low patronage; 3=Influence of other competitors, 4=Storage-related issues; 5=Others (specify) _______________________________________________ ___________________________________________) 13. What do you think is needed to boost your sales? (1=______________________________________________________________________ (2=______________________________________________________________________ 14. Do you have any physical or technical barriers to selling in any place of your choice in this area? __ (1=Government restriction; 2=Restriction by market as- sociation’s leader; 3=Restriction by other custom or tradition; 4=Others (speci- fy)_________________________________________________________________________________) 15. Is there any quality demanded by customers of your products that you think you are not satisfying pres- ently? ___ (1=Taste; 2=Tuber flesh color; 3= Size; 4= Price; 5=Storability; 6=Cooking time) 53 16. What is the general perception of your customers about your yam products? __ (1=Satisfied; 2=Not satisfied; 3= Indifferent) 17. In which month of the year is the marketing of this yam product high? __________________ And why? ___________________________________________________________________________________ Part H. Stress Incidence In your opinion, what are the major constraints to the yam production? (Please rank according to importance with 1 = highest rank, please be specific and use codes in the table) Constraints Rank Now 5 years ago Kindly pick among the following key constraints, the major 5 in yam pro- 1st duction? (Please rank them) 2nd 1-Planting material; 2-Pests & diseases;3-Water logging; 4-Drought; 5-Rodents; 6-Low soil 3rd fertility; 4th 7-Shortage of staking material; 8-Inadequate input supply; 9-Inadequate storage facilities; 5th 10-Land shortage; 11-High cost of labor; 12-Lack of new varieties; 13-Lack of credit; 14. Access to markets; 15-other (specify) ________________________? Part I. Transfer and Other Sources of Income Last Year Who earned/received? Sources of Total income (cash & in-kind) income Cash Total income Use first name used before Payment in kind (Naira/Cedi) Use Codes A (Naira/Cedi) Cash equivalent 1 2 3 4 5= 3+4 54 Codes A 1. Rented/sharecropped out land 8. Pension income 16. Sale of crop residues 2. Rented out oxen for ploughing 9. Drought/flood relief 17. Quarrying stones 3. Salaried employment 10. Safety net or food for work 18. Rent al property (other than land and oxen) 4. Farm labor wages 11. Remittances (sent from non-resident family and relatives living elsewhere) 19. Interest from deposits 5. Non-farm labor wages 12. Marriage gifts 20. Social cash transfer 6. Non-farm agribusiness income (e.g., grain milling/trading) 13. Sales of firewood/charcoal 21. Other, specify ……………..…….... 7. Other business NET income 14. Brick making (shops, 15. Poles from own and communal forests trade, tailor, sales of beverages, etc.) 55 56 Part J. Household Expenditure (Here, the person/s involved in purchases should be the principal respondent/s) Section 1: Food consumption Unit (e.g., Bought in the last 12 months No. kg, liter, Frequency of buy- Average quan- Item packet, ing (e.g., once/ tity each time Total quan- Average price/ Total cost bundle, year, twice/year, (e.g., 2 kg; 4 tity /year unit of purchase number) etc.) bundles, etc.) (Naira/Cedi) (Naira/Cedi) 1 2 3 4 5 6=4x5 7 8=6x7 Staple foods 1 Seed yam 2 Ware yam 3 Dried yam products 4 Maize 5 Wheat 6 Barley 7 Rice 8 Sorghum 9 Millet 10 Cassava 11 Potatoes 12 Sweet potato 13 Beans 14 Cowpea 15 Groundnut 16 Soybean 17 Pigeon pea 18 Banana 19 Plantain 20 Egusi/Melon 21 Other, specify..... Beverages and drinks 22 Tea (leaves) 23 Tea (liquid) 24 Coffee (powder) 25 Coffee (liquid) 26 Soft drinks 27 Juices 28 Local beer 29 Bottled/clear beer 30 Wine/Akpetshie/Dry Gin or hard drinks 31 Drinking water 32 Coffee beans 33 Opaque beer (chibuku/ pito) Section 1. Food consumption (cont’d) Unit Bought in the last 12 months No. (e.g., kg, Frequency of Average Total cost of Item liter, packet, buying (e.g., Average Total quan- price/unit purchase bundle, once/year, quantity number) twice/year) each time tity/ year (Naira/ Cedi) (Naira/Cedi) 1 2 3 4 5 6=4x5 7 8=6x7 Fruits 34 Orange 35 Mango 36 Pawpaw 37 Pineapple 38 Banana (ripe) 39 Apple 40 Guava 41 Coconut 42 Sugarcane 43 Other. .... Meat & other products 44 Beef 45 Goat meat 46 Mutton 47 Pork 48 Chicken 49 Turkey 50 Duck 51 Bush meat 52 Fish 53 Eggs 54 Milk 55 Cheese/Ghee 56 Butter 57 Yoghurt 58 Honey 59 Other. .... Vegetables 60 Tomato 61 Onion 62 Cabbage 63 Spinach 64 Kale 65 Carrot 66 Okra 67 Pumpkin 68 Egg plant 69 Cucumber 70 Pepper 71 Garlic 57 Section 1. Food consumption (cont’d) Unit Bought in the last 12 months No. (e.g., kg, Frequency Item liter, packet, of buying Average Total Average Total cost of bundle, num- (e.g., once/ quantity quantity/ price/ unit purchase ber) year, twice/ each time year year, etc.) (Naira/Cedi) (Naira/Cedi) 1 2 3 4 5 6=4x5 7 8=6x7 Fats, oils, sweeteners, snacks and others 72 Cooking fat 73 Margarine 74 Groundnut oil 75 Coconut oil 76 Bread 77 Biscuits 78 Popcorn 79 Cashew nuts 80 Sugar 81 Salt 82 Chocolate 83 Curry 84 Ginger 85 Macadamia nuts 58 Section 2. Expenditure on non-food items in the last 12 months Unit Frequency of purchase Average Total cost of purchase No. Expense Item (e.g. kg,, Average Total price/ unit liter, packet, (e.g., once/ quantity quantity bundle, year, twice/ each time year (Naira/ (Naira/ number) year,) Cedi) Cedi) 1 2 3 4 5 6 7 8=6x7 1 Clothing 2 Shoes 3 Blankets 4 Bed sheets 5 Soap/washing products 6 Electricity 7 Fuelwood 8 Charcoal 9 Kerosene 10 Batteries 11 School fees 12 School books and supplies 13 Health care 14 Grain milling 15 Land tax 16 Church contributions 17 Dowry 18 Membership fees 19 House building/construction 20 Guard/security 21 Newspapers, magazines, etc. 22 Travel expenses 23 Mobile phone air time (voucher) 24 Radio/TV service charge 25 Payment for extension services 26 Kitchen utensils 27 Personal care (toothpaste, nail, etc) 28 Furniture (tables, chairs, beds, etc.) 29 Home repairs 30 Purchase of bicycle, motorcycle, etc 31 Repairs for vehicles, bicycles, etc. 32 Petrol and engine oil for cars 33 House rent 34 Utility bills (water, telephone, etc.) 35 Cigarettes, tobacco, etc. 36 Remittances paid 37 Boxes of matches 38 Debt payments 39 Payment for land rent in cash 40 Other, specify............... 59 Part K. Access to Capital and Support Services Section 1. Household credit needs and sources during last cropping season. If the credit is in non-cash form, indicate the cash equivalent or value. If No in If Yes in mn 2, If No in col- Needed column colu umn 4, then If Yes in column 4 Activity credit? 2, then then did Have you why? you get what was the main Source of How much Codes A it? reason? credit, did you get? repaid the loan? Codes B Codes A (codes C) Codes D (Naira/Cedi) Codes A 1 2 3 4 5 6 7 8 1.Buying local seeds 2. Buying new seeds 3. Buying fertilizer 4. Buying herbicide/pesticides 5. Buying farm implements 6. Buying livestock 7. Investing in irrigation system 8. Non-farm business or trade 9. Buying food 10. Medical expenses 11. School fees 12. Others: ___________________ Codes A Codes B Codes C Codes D 0. No 1. Not cash con- 0. No reason 4. Expected to be re- 7. Lenders don’t provide 1. Money lender 6. Savings and 1. Yes strained 1. Borrowing is jected, so did not try it the amount needed 2. Farmer group/ Credit 2. Activity is not risky 5. I have no assets for 8. No credit association coop. 7. Relative/friend profitable 2. Interest rate collateral available 3. Merry-go- /neighbor 3. Never thought of is high 6. No money lenders 9. Not available on time round 8. Other, this investment 3. Too much in this area for this 10. Other, specify……… 4. Microfinance specify………….. 4. Other, specify....... paperwork/ purpose 5. Bank procedures Section 2. Access to extension/information services Did you receive training umn 2, or information on […...] If Yes in column 2, main If Yes in col number of contacts Type of service during the last cropping source of information/train- during the season? season? ing? (days/year) (Codes A) (Codes B) 1. New varieties of yam 2. New varieties of other crops 3. Pest and disease control - yam 4. Pest and disease control – other crops 5. Soil and water management- yam 6. Soil and water management-other crops 6. Crop rotation including yam 7. Output markets and prices 8. Input markets and prices 9. Livestock production 10. Family health/planning 11. Sanitation 12. Food processing Codes A Codes B 0. No 1. Government’s extension service 4. NGOs 7. Farmers’ 10. Mobile phone 13.Traders / Agro- 1. Yes 2. Farmers’ Coops or groups 5. Private company field school 11. Town hall meetings dealers 3. Neighbor/relative farmers 6. Research center 8. Radio/TV 12. Farmers’ training 14. Other specify........ 9. Newspaper center 60 Annex 1. Crop Codes Roots/Tubers/Banana/Plantain Cereals Grain legumes,Oil seeds, & Spices Industrial Tree Crops 1 Yam 11 Maize 21 Cowpea 51 Cocoa 2 Cassava 12 Rice 22 Pigeon pea 52 Coffee 3 Cocoyam 13 Sorghum 23 Groundnut 53 Oil palm 4 Sweet potato 14 Millet 24 Bambara nut 54 Coconut 5 Irish potato 15 Wheat 25 Cotton seed 55 Rubber 6 Plantain 16 Beniseed 26 Soybean 56 Cola nut 7 Cooking banana 17 Guinea corn 27 Egusi/ 57 Cashew 8 Frafra potato 18 Others……… 28 Melon 58 Citrus 9 Others……………. 29 Irvingia 59 Mango 30 Sesame seeds 60 Other…………… 31 Calabash Other industrial crops 32 Ginger 61 Sugarcane 33 Green grain 62 Sisal 39 Others ………….. 63 Tobacco 40 Vegetables 64 Kenaf 65 Cotton 66 Other: _____________ 61 4. Community level questionnaire International Institute of Tropical Agriculture (IITA) Yam Improvement for Income and Food Security in West Africa (YIIFSWA) COMMUNITY LEVEL QUESTIONNAIRE Nigeria and Ghana Part A. Interview Background 1. Country No. ______ (0 =Ghana; 1 = Nigeria) 2. State/Region: ______________________________________ 3. LGA/District: ______________________ 4. Community/Village name _________________________________ 5. Survey date: Day _____; Mth _____ ; 20 ______ 6. State of road from main city to community: _____ (Use codes below of roads status) 1 Tarmac, easily motorable in all seasons; 4 Path, easily passable in all seasons; 7 Dirt road, easily motorablein all seasons; 2 Tarmac, poorly motorable in all seasons; 5 Path, barely passable in all seasons; 8 Dirt road, barely motorable in all seasons; 3 Tarmac, not motorable in all seasons; 6 Path, not passable in all seasons; 9 Dirt road, not motorable; 10 River or stream. 7. No. of people: _________ interviewed, comprising ________ men and __________women 8. GPS readings Latitude (N/S) Longitude (W/E) Altitude in meters . 0 . 0 Part B. Crops Grown 1. What are the main crops grown in this community? (Rank first = most important) Crops* ranked by Overall importance Land area allocated Volume of sales Quantity consumed 1 1 1 1 2 2 2 2 3 3 3 3 * Roots tubers and plantain 1 Yam; 2 Cassava; 3 Cocoyam; 4 Sweet potato; 5 Irish potato; 6 Plantain; 7 Cooking banana; 8 Other roots/tubers Cereals 11 Maize; 12 Rice; 13 Sorghum; 14 Millet; 15 Wheat; 16 Finger millet; 17 Other cereals Grain legumes, oil seeds and vegetables 21 Cowpea; 22 Pigeon pea; 23 Groundnut; 24 Bambara nut; 25 Cotton seed; 26 Other beans/peas; 27Egusi/melon; 29 Sesame seed; 30 Calabash; 31 Ginger; 32 Sunflower; 33 Beniseed; 34 Tea; 35 Other legumes/oils; 40 Vegetables 2. Do you know anyone producing only seed yam? ____ (1 = Yes; 0 = No) 2.a. If YES, are they many in this community? ___ (1 = Yes; 0 = No) Number? _____ male & ______ female 3. Do you have in this village any yam variety with extraordinary qualities? ___ (1 = Yes; 0 = No) 62 3a. If Yes, what are they? Variety1 Name ________________________; Qualities ______________________________________ Variety2 Name ________________________; Qualities ______________________________________ 4. Have you faced scarcity of certain good yam varieties which existed in the past? __ (1=Yes; 0=No) If Yes, which ones?___________________________________________________________________ 5. Which other good yam varieties disappeared completely? __________________________________ And why? _________________________________________________________________________ 6. What has been the trend in yam production in the last 20 years? ___ (1 = Decreasing? 2 = No change? 3 = Increasing?) Why? Explain _______________________________________________________________________________ 7. What has been the trend in yam production in the last 4 years with YIIFSWA interventions? ___ (1 = Decreas- ing? 2 = No change? 3 = Increasing?) 8. Yam production goals. What is the most important objective for growing yam in this community? ___ (1 = Sale; 2 = Food; 3 = Other, specify: ___________________________) 9. Source of hired labor: Where does the hired labor in this community come mostly from? ___ (1=Within the com- munity; 2=Neighboring community in the area; 3=Community far away (in other regions); 4=Nearest town; 5= Neighbor- ing countries; 6=Not known) Part C. Risk Sources & Infrastructure 1. What are the major problems in the production of yam in the community? 1. ____________________________________________________________________ 2. ____________________________________________________________________ 3. ____________________________________________________________________ 2. Where do farmers sell yam mostly? ___ (1 = Farm gate; 2 = Village market; 3 = Other market), specify: ___________________________) 3. By what (most common means) do you carry yam to market? ____ (1 = Head load; 2 = Bicycle; 3 = Barrow/Cart, 4=Lorry/Pickup/tractor/trailer; 5= Animal; 6=Motorcycle; 7=Other) specify: ______________________________) 4. Where is your market located? ______ (1=inside the village; 2=Outside the village) 5. What is the frequency of market days? Every _______ days 5.1 Rank by volume traded, the people who buy ware yam in this market. People who buy (Rank 1=highest) Consumers from this or nearby community? ___ Consumers from far away? ___ Small traders from this and nearby villages? ___ Small traders from far away? ___ Big traders from far away with lorries? ___ 63 5.2 Rank by volume traded, the people who buy seed yam in this market. People who buy (1=highest) Consumers from this or nearby community? ___ Consumers from far away? ___ Small traders from this and nearby villages ___ Small traders from far away? ___ Big traders from far away with lorries ___ 5.3 Rank by volume traded, the people who sell ware yam in this market. People who sell (Rank 1=highest) Farmers themselves? ___ Traders from this and nearby community? ___ Traders from far away? ___ 5.4 Rank, by volume traded, the people who sell seed yam in this market. People who sell (Rank 1=highest) Farmers themselves? ___ Traders from this and nearby community? ___ Traders from far away? ___ 6. How many vehicles (lorries) come into the market per market day? ____________ 64 5. Field level questionnaire International Institute of Tropical Agriculture (IITA) Yam Improvement for Income and Food Security in West Africa (YIIFSWA) FIELD LEVEL QUESTIONNAIRE Nigeria and Ghana Part A. Interview Background 1. Survey date: Day ____ ;Mth _____ ; 20_____ Name of respondent: ______________________________ 2. Field location distance from residence: ____________ (in minimum walking time) 3. GPS readings of the yam field Latitude (N/S) Longitude (W/E) Altitude in meters . 0 . 0 4. How old is your yam field? 4.a How many people have worked on this field? _____ And how many years has this land been put into use? __________ 4.b How many years have you been using this same land? _______ 5. Has your community benefited from YIIFSWA project? _____ (1=Yes; 0=No) 6. If Yes from Question 5, indicate how. ____________ (1= Training in minisett technique; 2= Training in vine cutting technologies; 3= Training in business plan development for pre-basic and basic seed producers; 4= Training in business plan development for seed producers; 5=Training in business plan development for yam producers; 6=Provided with seed tubers; 7=Provided with QDS/pre-basic/basic materials for seed production; 8=Provided with plantlets; 9=Training on high-ratio seed yam propagation techniques; 10=Pro- duction of certified seed yam; 11=Training in seed yam quality control and certification; 12=Improved yam storage facilities; 13= Others (specify: __________________________________) 7. Have you in this community put the benefit gained into use? ____ (1= Yes; 0= No) 8. If Yes to question 7 above, how many have benefited from your community? _________ 9. If Yes to question 7 above,, how has it helped you in this community?______ (1= Yam production increased; 2=Area of yam produced increased; 3= Output per unit area increased; 4= Reduced losses; 5= Improved quality and safety of processed products; 6=Application of quality management protocols system in seed yam production; 7= Others (specify: ____________________________________________________) 10. In which year did you start benefiting from YIIFSWA project in this community? _____ (1=2012, 2=2013; 3=2014; 4=2015) 11. How satisfied are you with the project’s benefit(s)? _____ (1= Very satisfied; 2= Satisfied; 3= Not satisfied) 12. If not satisfied with the project benefit(s), why not? ____ (1= Lack/shortage/unavailability of input; 2=Not enough time for learning; 3= Not convinced; 4= Poor quality of planting material received; Others, specify: __________________________________________________________________________________________) Part B. Yam & Other Crops Grown 1 How much did you spend on the seed yam you purchased? _________________ Naira/Cedis 2 Did you sell any of the seed yam? _____ (1=Yes; 0=No) 3 If Yes to question 2, for how much did you sell the own produced seed yam? ____________ Naira/Cedis 4 If No to question 2, what is the main reason for not selling? ___ (1=Not producing seed yam, 2=Not enough, 3=Not profitable, 4=Fear of bad luck from seed yam sale, 5=Fear of losing production supremacy, 6=Others,)specify: ________ _____________________________________) 65 Part C. Labor Input Use For Yam Grown 1 Land clearing 1.1Who did most of the land clearing in this field? ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other, specify: _______________________________________) 1.2 If mostly by men: 1.2a How many men working full time would clear the entire field in one day? ________ men 1.2b What was the wage rate/man/day for the land clearing? ____________ Naira/Cedis 1.3 If mostly by women: 1.3a How many women working full time would clear the entire field in one day? ____women 1.3b What was the wage rate/woman/day for the land clearing? ____________ Naira/Cedis 1.4 If mostly by children < 15 years: 1.4a How many children <15 working full time would clear the entire field in one day? _____children 1.4b What was the wage rate/child/day for the land clearing? ____________ Naira/Cedis 1.5 How much of the entire land clearing labor for this field was hired and how much was family? _____ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 1.6 Was any of the land in this field mechanized? ___ (1 = Yes; 0 = No) 1.6a If Yes, was it mechanized, in full or in part? ____ (1= Mechanized fully; 2 = Mechanized partly) Type of mechanization: _______________________ 1.6b How much was paid for the mechanization? __________________ Naira/Cedis 1.7 Did you apply herbicide? ___ (1 = Yes; 0 = No); If Yes, cost ____________ Naira/Cedis 2 Seedbed Preparation 2.1Who did most of the seedbed preparation in this field? ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other, specify: ______________________________________) 2.2 How much of the entire seedbed preparation labor for this field was hired and how much was family? _____ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 2.2b Was any of the seedbed preparation in this field mechanized? ___ (1 = Yes; 0 = No) 2.2c If mechanized, in full or in part: ____ (1= Mechanized fully; 2 = Mechanized partly) Type of mechanization: _______________________ 2.3 If mostly by men: 2.3a How many men working full time would prep. the seedbed in the entire field in one day? ________ men 2.3b What was the wage rate/man/day for the seedbed preparation? ____________ Naira/Cedis 66 2.4 If mostly by women: 2.4a How many women working full time would prepare the seedbed in the entire field in one day? ____women 2.4b What was the wage rate/woman/day for the seedbed preparation? ____________ Naira/Cedis 2.5 If mostly by children < 15 years: 2.5a How many children <15 working full time would prepare the seedbed in the entire field in one day? _____children 2.5b What was the wage rate/child/day for the seedbed preparation? ____________ Naira/Cedis 2.6 If mechanized, in full or in part? 2.6a How much was paid? __________________ Naira/Cedis 3 Planting 3.1Who did most of the planting in this field? ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other): specify:_________________________________________) 3.2 How much of the entire planting labor for this field was hired and how much was family? _____ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 3.3 If mostly by men: 3.3a How many men working full time would plant yam in the entire field in one day? ________ men 3.3bWhat was the wage rate/man/day for the planting? ____________ Naira/Cedis 3.4 If mostly by women: 3.4a How many women working full time would plant yam in the entire field in one day? ____women 3.4bWhat was the wage rate/woman/day for the planting? ____________ Naira/Cedis 3.5 If mostly by children< 15 years: 3.5a How many children <15 working full time would plant yam in the entire field in one day? _____children 3.5b What was the wage rate/child/day for the planting? ____________ Naira/Cedis 4 Weeding 4.1Who did most of the different weedings done in this field? 4.1a For Weeding 1: ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other): specify:_____________________________________________) 4.1b For Weeding 2: ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other): specify:_____________________________________________) 4.1c For Weeding 3: ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other, specify:_____________________________________________) 67 4.2 How much of the entire weeding labor for this field for each weeding was hired and how much was family? 4.2a For Weeding 1: ___ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 4.2b For Weeding 2: ___ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 4.2c For Weeding 3: ___ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 4.3 Did you weed with herbicide? ___ (1 = Yes; 2 = No); If Yes, what is the cost? _________ Naira/Cedis 4.4 For the weedings done mostly by men: 4.4a How many men working full time would weed the entire field in one day? Weeding 1 Weeding 2 Weeding 3 _______ men ______ men ______ men 4.4b what was the wage rate/man/day for weeding? Weeding 1 Weeding 2 Weeding 3 _______ Naira/Cedis ______ Naira/Cedis ______ Naira/Cedis 4.5 For the weedings done mostly by women: 4.5a How many women working full time would weed the entire field in one day? Weeding 1 Weeding 2 Weeding 3 _______ women ______ women ______ women 4.5b What was the wage rate/woman/day for weeding? Weeding 1 Weeding 2 Weeding 3 _______ Naira/Cedis ______ Naira/Cedis ______ 4.6 For the weedings done mostly by children: 4.6a How many children working full time would weed the entire field in one day? Weeding 1 Weeding 2 Weeding 3 _______ children ______ children ______ children 4.6b What was the wage rate/child/day for weeding? Weeding 1 Weeding 2 Weeding 3 _______ Naira/Cedis ______ Naira/Cedis ______ 5 Harvesting 5.1 Who did (would do) most of the harvesting in this field? ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other): specify: _______________) 5.2 How much of the entire harvesting labor for this field was hired and how much was family? _____ (1=All family; 2=Mostly family; 3=Hired/family equally; 4=Mostly hired; 5=All hired) 68 5.3 If mostly by men: 5.3a How many men working full time would harvest yam in the entire field in one day? ________ men 5.3b What was the wage rate/man/day for the harvesting? ____________ Naira/Cedis 5.4 If mostly by women: 5.4a How many women working full time would harvest yam in the entire field in one day? ____women 5.4b What was the wage rate/woman/day for the harvesting? ____________ Naira/Cedis 5.5 If mostly by children< 15 years: 5.5a How many children <15 working full time would harvest yam in the entire field in one day? _____children 5.5b What was the wage rate/child/day for the harvesting? ____________ Naira/Cedis 6 Transportation 6.1 Where did (would) you carry most of the yam harvested? ___ (1 = Home; 2 = Market; 3=Other): specify: _________________________) 6.2 By what most common means did (would) you carry the yam harvested? ____ (1 = Head load; 2 =Bicycle; 3 = Barrow/Cart, 4=Lorry/Pickup/tractor/trailer; 5= Animal; 6=Motorcycle; 7=Other): specify: _________________________________________) 6.3 If mostly by head load (1) 6.3(a) Who did most of the carrying for the yam harvested? ___ (1=Men mostly; 2=Women mostly; 3=Both equally; 4=Children < 15; 5=Other): specify:____________________________________) 6.3(b) How many of the people would carry all the yam in one day? ___________ people. Part D. Non-Labor Input Use for Yam Grown 1 Have you used stakes? ___ (1 = Yes; 0 = No). If Yes, how much did you spend on stakes?_______ Naira/Cedis 2 Have you used chemical fertilizer? ___ (1 = Yes; 0 = No). If Yes, how much did you spend on it? ___________ Naira/Cedis 3 Have you used other chemical 1? (specify: _______________________________________)? ___ (1 = Yes; 0 = No). If Yes , how much did you spend onit?_______________ Naira/Cedis 69 4 Have you used other chemical 2 (specify: _______________________________________)? ___ (1 = Yes; 0 = No). If Yes, how much did you spend on it? _______________ Naira/Cedis Part E. Tenurial Arrangements 1 Who owns the yam in this field? ___ (1=Whole family; 2=Man or husband; 3=Woman or wife; 4=Son; 5=Daughter; 6=Other): specify: __________________________________) 2 How was this land acquired for use in producing yam? ___ (1=Inherited; 2*=Loaned or Rented; 3=Borrowed; 4=Purchased; 5=Allocated by; 6=Other):, specify: ___________________________) * if option 2 from ques- tion 2, indicate the mode of payment: ___ (1=Cash; 2=Kind; 3=Sharecrop; 4=Other _____) 3 If this field or land has been inherited, from who was it inherited? ___ (1=Father’s (Husband’s) family; 2=Mother’s (wife’s) family) Part F. Harvests & Uses of Yam Output 1 Total field area: __________ m2. 2 Yam field sample sizes: Size 1 ____________ sqm.; Size 2 ____________ sqm.; Size 3 ____________ m2. 3 Number of stands/points: 1_________stands; 2 _____________ stands; 3 ______________ stands 4 Yield, Number of tubers per stand/point: 1 __________________; 2 _________________; 3 _____________________ 5 Weight of tubers per stand/point: 1 __________total kg; 2 __________ total kg; 3 __________total kg 6 Is the field milked? ___ (1 = Yes; 0 = No) 7 If you consider your yam harvest to be 10 parts, how many parts have you harvested early (July to September) and how many parts will you harvest later? Harvest early _______ parts out of 10 Harvest later _______ parts out of 10 ________________ Total 10 8 If you divide your expected yam output from this field into 10 parts, how many parts will you sell and how many parts will you use at home and as seed yam? Sell _______ parts out of 10 Home use _______ parts out of 10 Used as seed yam _______ parts out of 10 ________________ Total 10 70 9 If you divide yam to sell from this field into 10 parts, how many parts will you sell now and how many parts will you store and sell later? Sell now _______ parts out of 10 Store to sell later _______ Parts out of 10; Store for how long? ______ months ________________ Total 10 Weighing records in kg Series 1 2 3 4 5 6 Total 1 2 3 Total 10. Observation, if any special way of producing yam from this field_____________________________________ _________________________________________________________________________________________ 71 YIIFSWA Working Paper Series 1. Yam Improvement for Income and Food Security in West Africa YIIFSWA Project Description 2. Seed Yam Production in an Aeroponics System: A Novel Technology 3. Yam: A Cash Crop in West Africa 4. Baseline Protocols. The Case of Yam Improvement for Income and Food Security in West Africa (YIIFSWA) Project 5. Working with farmers to produce clean seed yams 6. Novelty, rapidity and quality in seed yam production: the case of Temporary Immersion Bioreactors 72