The conference agenda provides an overview and details of sessions. In order to view sessions on a specific day or for a certain room, please select an appropriate date or room link. You may also select a session to explore available abstracts and download papers and presentations.
13-09: The Root of the Measure: Methodological Experiment
For more information or signing up, please contact email@example.com
The Root of the Measure: Methodological Experiment on Cassava Production and Variety Identification
1World Bank; 2Lilongwe University of Agriculture and Natural Resources
This paper reports results from a randomized methodological experiment conducted across 5 top-cassava districts and 45 enumeration areas in Malawi from July 2015 - August 2016. In each EA, sampled cassava households were randomly assigned to harvest diaries or two 6-month recall surveys or a single 12-month recall survey for collecting information on cassava production. The diary sample was further randomized into equal halves that received either in-person visits by enumerators or mobile phone calls to monitor diary completion. Multivariate analyses will estimate the impact on annual household cassava production estimates stemming from 3 treatment arms in comparison to what is often touted as the gold standard: diary with in-person visits. Inter-arm relative accuracy and cost estimates will be provided, with a focus on exploring the feasibility of conducting mobile phone surveys. On objective cassava variety identification, similar analyses will be undertaken for the use of (i) farmer-reported variety name, and (ii) photo library of morphological descriptors, vis-à-vis DNA fingerprinting of cassava leaf samples obtained from one randomly-selected plot for all 1,260 households across the treatment arms, in which crop cutting was also conducted to provide an upper-bound benchmark for production and productivity measurement.