How to make farming and agricultural extension more nutrition-sensitive: evidence from a randomised controlled trial in Kenya

How can experiments play a greater role in public policy? Twelve proposals from an economic model of scaling

“…This is an area of inquiry that policymakers are well aware of, and one that at least on traditional individual-specific observablesgender, race, agethey keep close tabs. Scholars have also made important inroads in these dimensions (Bell & Stuart, 2016;Stuart et al, 2018), lending to important insights into heterogeneity (Heckman et al, 1998b) and the effects of non-random attrition (see, e.g., Ogutu et al, 2018). Yet, there are in many cases invaluable pieces of information that the researcher has that the policymaker might not have at their disposal or that it might be difficult to obtain.…”

“…In their study of take-up of biofortified crops by farmers, Ogutu et al (2018) acknowledge the possibility that non-random attrition may affect their treatment effect estimate because the control group and one of three treatment arms had higher attrition rates than the other two. If characteristics that influenced the program impact were related to characteristics that led subjects to attrit, the measured treatment effect would be a poor representation of the actual treatment effect.…”

How Can Experiments Play a Greater Role in Public Policy? 12 Proposals from an Economic Model of Scaling

“…This is an area of inquiry that policymakers are well aware of, and one that at least on traditional individual-specific observables-gender, race, age-they keep close tabs. Scholars have also made important inroads in these dimensions too (Bell and Stuart, 2016;Stuart et al, 2018), lending important insights into heterogeneity and the effects of non-random attrition (see, e.g., Ogutu et al, 2018). Yet, there are in many cases invaluable pieces of information that the researcher has that the policymaker might not have at her disposal or that it might be difficult to obtain.…”

“…In their study of take-up of biofortified crops by farmers, Ogutu et al (2018) acknowledge the possibility that non-random attrition may affect their treatment effect estimate because the control group and one of three treatment arms had higher attrition rates than the other two. If characteristics that influenced program impact were related to characteristics that led subjects to attrit, measured treatment effect would be a poor representation of the actual treatment effect.…”

Making Cash Crop Value Chains Nutrition-Sensitive: Evidence from a Quasi-Experiment in Rural Sierra Leone

“…In the light of increasing recognition that hunger and malnutrition need to be fought in multiple fronts (Nisbett et al, 2016;World Bank, 2007), there has been a growing call and support for integrated, nutrition-sensitive interventions by governments, donors, and development practitioners (Bhutta et al, 2013;Ruel & Alderman, 2013;Ruel, Quisumbing, & Balagamwala, 2018). One of such approaches, with significant promise to address these problems in smallholder communities, is the nutrition-sensitive value chain (NSVC) model, which combines agricultural and nutrition-related interventions to promote both good agricultural practices and good nutritional practices along value chains (Allen & de Brauw, 2018;De la Pena & Garrett, 2018;Gelli et al, 2015;Hawkes & Ruel, 2012;Ruel & Alderman, 2013) A growing body of research has demonstrated that such nutrition-sensitive interventions, mainly in food crop and livestock value chains, have improved production of, access to, and intakes of nutrient-rich foods; enhanced women's status; reduced morbidity and improved some dimensions of nutritional status of household members (Kumar et al, 2018;Leroy et al, 2016;Nisbett et al, 2016;Ogutu et al, 2018;Olney, Pedehombga, Ruel, & Dillon, 2015;Rosenberg et al, 2018). Empirical evidence is, however, lacking on the impacts of these integrated approaches in non-food, cash crop sectors, which are riddled with food insecurity and malnutrition (De Vries et al, 2012, 2013b, 2013aFreeman et al, 2014).…”