Impact assessment based on self-reported attribution in complex contexts of rural livelihood transformations in Africa.

发展机构内部对影响证据的需求越来越大，既是为了支持自己的学习，也是为了对其他利益攸关方负责。内部监测和评估系统是这类评估的重要基础，但人们普遍认识到，定期的独立影响评估可以有效地补充这些系统。然而，对MACH IA的可靠性和/或成本效益的怀疑仍然存在，促使关于替代方法的持续试验和学术辩论。拟议的研究解决了一个特别困难的问题，即如何评估旨在使非洲农村转型复杂进程中的贫穷男女受益的发展活动的影响。所处理的核心问题是如何以可靠、及时和具有成本效益的方式将变化的具体干预措施赋予不同的利益攸关方，同时又不扭曲或限制正在评估的发展活动。虽然影响归属问题在发展实践中普遍存在，但这项研究将特别侧重于两个非政府组织的工作：非洲自助组织(SHA)和非洲农场组织(FA)，以及它们在马拉维和埃塞俄比亚正在开展的活动。更具体地说，这项研究将评估在这种情况下进行严格和具有成本效益的影响评估的范围，其依据是：(A)对家庭一级的经济安全进行系统监测；(B)进行深入访谈，以促使受访者自我报告其归属。这种方法可以与影响评估的参与性方法以及那些依赖于使用随机对照试验或准实验方法的统计推断归因的方法进行对比。拟议的研究包括三个互补的方面。第一部分：四项基线研究(两项在马拉维，两项在埃塞俄比亚)，随后在非政府组织发展证据(EFD)的专家技术支持下，由民政事务局和财务顾问进行了两轮计划监测研究。这将借鉴和加强EFD在开发综合家庭方法(IHM)方面的开创性工作，以衡量和模拟家庭水平、现金收入和食品消费的恢复情况。第二轮：独立顾问的两轮影响研究，使用定性的个人影响方案(QUIP)，将第一轮监测与基于自我报告的归因方法的半结构化访谈相结合。第三轮：两轮元评估，通过这两轮元评估，将审查通过第一轮和第二轮获得的证据的质量，并收集其他证据，说明不同利益相关者如何解释和使用证据，以及如何影响他们的工作关系。这里特别关注的是通过Strand 2获得的证据在从Strand 1获得的证据之外又增加了什么，以及这是否证明额外的费用是合理的。因此，这项研究将有助于改善支持改善发展干预措施的反馈循环，以支持非洲农村生计的转变，并具有支持更广泛援助有效性的潜在方法效益。除了研究的核心方法目标外，研究还将探讨两个参与研究的非政府组织在埃塞俄比亚和马拉维的工作中面临的紧迫和实质性问题，这些问题也与开展类似活动的其他组织有关。更具体地说，卫生局和足协正在寻求提高他们对哪种干预措施组合在改善两国贫困农村家庭的经济和粮食安全方面最有效的理解。要调查的活动包括促进改良种子和耕作方法、畜牧业、供应链升级、自然资源管理、小额信贷和非农业活动多样化。

There is a growing demand for impact evidence within development agencies, both to support their own learning and to account responsibly to other stakeholders. Internal M&E systems are an important foundation for such assessment but it is widely recognised that they can usefully be complemented by periodic independent impact assessment (IA). However, doubts remain over the reliability and/or cost-effectiveness of much IA, prompting ongoing experimentation and academic debate over alternative approaches.The proposed research addresses the particularly difficult issue of how to assess the impact of development activities intended to benefit poor men and women caught up in complex processes of rural transformation in Africa. The core issue addressed is how to attribute change specific interventions in a way that is reliable, timely and cost-effective to different stakeholders, without at the same time distorting or constraining the development activity being assessed. While the impact attribution problem is pervasive in development practice, the research will be focus particularly on the work of two NGOs: Self Help Africa (SHA) and Farm Africa (FA), and their ongoing activities in Malawi and Ethiopia. More specifically the research will assess scope in such contexts for rigorous and cost-effective impact assessment based on (a) systematic monitoring of economic security at the household level combined (b) with in-depth interviewing designed to elicit self-reported attribution by respondents. This approach can be contrasted with both participatory approaches to impact assessment, and those that rely on statistically inferred attribution using either randomised control trials or quasi-experimental methods. The proposed research comprises three complementary strands. Strand 1: four baseline studies (two in Malawi, two in Ethiopia) followed by two rounds of programme monitoring studies conducted by the SHA and FA with specialist technical support from the NGO Evidence for Development (EfD). This will draw upon and enhance EfD's pioneering work in developing the Integrated Household Method (IHM) for measuring and modelling the reslience of household level cash income and food consumption. Strand 2: two rounds of impact studies by independent consultants, using a qualitative individual impact protocol (QUIP) that combines Strand 1 monitoring with semi-structured interviews based on a self-reported attribution methodology.Strand 3: two rounds of meta-evaluation, through which the quality of evidence obtained through Strands 1 and 2 will be reviewed, and additional evidence collectied of how it was interpreted and utilised by different stakeholders, and how it affected their working relationships. A particular focus here is on what evidence obtained through Strand 2 added to that obtained from Strand 1 and whether this justified the extra cost.The research will thereby contribute to improving feedback loops in support of improved development interventions to support rural livelihood transformation in Africa, with potential methodological benefits in support of aid effectiveness more generally. In addition to the core methodological goals of the research, the research will also address immediate and substantive questions facing the two participating NGOs in their work in Ethiopia and Malawi, relevant also to other organisations carrying out similar activities. More specifically, SHA and FA are seeking to improve their understanding of what combination of interventions are most effective in improving the economic and food security of poor rural households in the two countries. Activities to be investigated include promoting improved seeds and cultivation practices, animal husbandry, supply chain upgrading, natural resource management, microfinance, and diversification into non-farm activities.