The tiny village of Narma Dih, off-grid in Bihar, India, was lit only by the full moon and the beam of a battery-powered pico projector. A makeshift screen hung on the outside wall of a modest dwelling. A clump of small children clung to each other and stared at the screen, transfixed. Behind them sat a circle of sari-clad women, equally absorbed. A few men stood in back. The object of their rapt attention? Not a Bollywood extravaganza, but a locally produced how-to video on seed preparation for okra cultivation.
I was in Narma Dih to get a first-hand look at Digital Green, which uses technology to accelerate the diffusion of agricultural innovations. The WDR 2016 is all about storing and sharing information, and that is at the heart of agricultural extension. There can be high returns to putting the right information in the right hands at the right time. This is especially true if you can show farmers ways of being more productive with their existing resources -- for instance, showing them how to intercrop, or to make better compost. But credibly transmitting this kind of information has always been difficult, labor-intensive and costly. Agricultural extension agents are typically assigned to serve an impossibly large number of farmers spread over a logistically daunting stretch of countryside. And the traditional form of information transmittal leaves something to be desired. In Bihar, the agents have travelled the back roads shlepping flipcharts, text-heavy and just plain heavy, one per topic. The flipcharts may not adequately convey new techniques to illiterate farmers, let alone give them confidence to try a risky new idea. Would you believe someone who told you that you could sow 90% fewer seeds while boosting your yield? (That's the promise of the system of crop intensification, whose diffusion is a goal of the Bank-supported Jeevika Project.)
Digital Green takes advantage of the drastic reduction in the costs of creating and transmitting information offered by three technologies: videocameras, pico projectors, and the internet. Local people film local farmers explaining in local dialects how they implement an agricultural innovation and the videos are then pooled on the internet. Then, resource people download and show seasonally appropriate videos to women's self-help groups like those in Narma Dih. The premise is that the home-grown videos convey the technical details of the technique, and do it with street cred.
From what I saw, the technology complements rather than substitutes for extension agents. Some of the farmers said that they were able to pick up and apply an innovation from a single viewing. But the village resource people also play a strong role in promoting adoptions (incentive payments are linked to adoptions). During the screening, they stop the video to stress key points and to answer questions. Later they are on call to remind farmers of forgotten details. Thus a tech-assisted resource person can effectively service far more clients than a traditional extension agent. Perhaps the resource people will be less critical in the future as smartphones and 4G service become cheaper and more prevalent. Then, farmers could download, store and replay videos and call a help center when needed.
As a monitoring and evaluation proponent, I am particularly fascinated by the rich monitoring data that Digital Green generates. It tracks which videos were shown where, and which ones resulted in adoptions. It's not automatic-- there's manual entry of details on dissemination and adoption -- but it feeds into a database which has a dashboard here. There's huge potential to figure out how characteristics of farmer, village, video, and presenter interact to determine adoption rates. Digital Green is just beginning to tap this potential, looking at where and why adoption rates are lower in some places than others -- see their bottleneck analysis here.
Altogether, it's an appealing package. And it's not terribly high-tech -- they manage to implement it in areas with spotty electricity and spottier internet access. But does it really work? Is it cost-effective? There's a randomized controlled trial underway, with results expected later this year. Stay tuned.