Recalling Extra Data: A Replication Study of Finding Missing Markets
复制了Ashraf等人关于肯尼亚出口作物推广干预的研究，发现原始结果总体稳健但部分系数较低，并探讨了理论变化、统计功效及回忆偏差等问题，对政策制定者和未来研究有参考价值。
We reexamine some of the strongest evidence supporting cash-crop-based development strategies by replicating Nara Ashraf, Xavier Giné and Dean Karlan’s ‘Finding missing markets (and a disturbing epilogue): Evidence from an export crop adoption and marketing intervention in Kenya’, published in the American Journal of Agricultural Economics. The original evaluation, of an agricultural export crop promotion intervention in Kenya, is one of the few impact evaluations exploring how agricultural commercialisation affects household outcomes. Our attempt to independently reconstruct the evaluation using the existing raw data finds the original results generally robust to replication, albeit with much lower coefficients on some of the main outcomes of interest. We explore the evaluation’s theory of change, focusing on the result that first-time export crop adopters benefit more from agricultural commercialisation than agricultural households that were already producing export crops. We also examine questions around adequate power requirements and potential recall and or courtesy bias within the analysis. Reproducing these original results is relevant both to encourage policymakers to use this evidence and to highlight knowledge gaps for future research.