Where YouTube Meets the Farm

Half of the children in India are chronically malnourished. So imagine the potential benefits if there were a simple way to increase the milk production from cows by, say, a quart per week Or imagine if there were a better way for small farmers to cultivate rice â€" the staple food of half the world â€" one that required no costly inputs, used less water, and substantially increased yields

Wouldn’t these things be development miracles

Well, they are not miracles; they are realistic possibilities. The problem is that most farmers in the developing world don’t know about them â€" or don’t know how to implement them successfully.

Consider: an aquatic fern called azolla, which can be readily cultivated and added to animal feed, can boost production of cows milk by 15 to 20 percent. An approach known as System of Rice Intensification (SRI), which involves transplanting rice saplings, spacing them in a grid, keeping the soil drier, and carefully weeding plots, can produce remarkable gains. SRI has been called one of the most important agricultural innovations of the past 50 years, yet it is employed by a fraction of farmers (pdf). And there are countless other opportunities: for instance, orange-fleshed sweet potatoes from a home garden can averVitamin A deficiency, which causes hundreds of thousands of children in the developing world to go blind or die each year.

One of the great paradoxes in today’s world is that information is so easy to transmit â€"few places on earth are beyond the reach of cellphones or televisions â€" and yet our efforts to get life-saving, livelihood-boosting information to people in a form that sticks, a form that will actually change behavior, are frequently disappointing.

That was a problem that gripped Rikin Gandhi, a young American-born software engineer, while he was working in Bangalore for Microsoft Research India seven years ago. Rikin was interested in how rural telecenters might be used to spread education and information about health and agriculture in remote areas. A colleague suggested he investigate the application of the Digital StudyHall model in rural Karnataka. Gandhi did just that â€" and his experience led to the creation of Digital Green, a platform and process for extending knowledge and influencing behavior that has seized the attention of many development experts.

India has 100,000 agricultural extension workers whose job is to translate research into useable information for the 60 percent of Indians who earn a livelihood from farming, many of whom cannot read and lack electricity. Historically, extension efforts have tended to bypass the poorest, however. “Only 40 percent of Indian farmers (pdf, p. 10) receive any kind of agricultural information, and only 6 percent are reached by the public sector extension agents,” observed Pier Paolo Ficarelli, an expert with 20 years of experience in the field. Even when extension agents reach villagers, adoption rates for new farming practices remain low. The question is: With work so vital to human well being, how can we do better

Gandhi, who had aspired to be an astronaut and studied aeronautical engineering at M.I.T., spent six months in villages in Karnataka experimenting with communication formats â€" posters, TV shows, locally made videos, public screenings, home screenings. He discovered that if he produced short (8- to 10-minute) videos that featured local farmers (both men and women, as most agricultural work in India is done by women) talking about their experiences and screened them with a facilitator who engaged a group in a discussion â€" an idea based on a teaching model pioneered by the Stanford researcher Jack Gibbons (pdf) â€" farmers were highly engaged. Not only did they sit through the videos and ask questions, many took up the practices. Kentaro Toyama, Gandhi’s boss at Microsoft, recalled: “He came back and said, ‘I think I have something that is much more effective than traditional agricultural extension.’ So we wantedo evaluate it.”

They set up a controlled trial, comparing Gandhi’s approach with a standard “Training and Visit-based” extension approach. Among 1,470 households in 16 villages, they found that it increased adoption of some agricultural practices sevenfold over control villages. The research indicated that the cost to get one farmer to adopt one practice dropped tenfold, from $38 with the traditional approach to $3.70 with the video-based model.

Gandhi and his colleagues decided that Digital Green should spin off as an N.G.O. The Gates Foundation provided support. Today, the organization has several funders, works in 2,000 villages in India, 100 in Ethiopia, and 50 in Ghana. Working with a variety of partners, it has produced 2,600 videos that have been viewed by 157,000 farmers. It reports that 41 percent of viewers in the last two months have adopted at least one practice. Gandhi now has 60 colleagues working with him and plans to be reaching 10,000 villages by 2015.

Rashmi Kanthi Nagaratna Hublimath, a community leader in Dharwad, Karnataka, participating in Digital Green video production training. While learning to use a digital camera, she captures an agricultural practice demonstrated by fellow members of her community.

What’s intriguing about Digital Green is how it uses videos to start public conversations and elicit leadership within communities. What’s unexpected is that it has been able to produce locally made videos in India’s regional languages at scale, something Gandhi notes is cheaper and faster than using professional filmmakers. “Digital Green’s approach is kind of the MOOC model turned upside down,” observes Richard Anderson, a professor in the department of computer science and engineering at the University of Washington, who is working with PATH to adapt its model to public health. “With a MOOC you have one centralized video reaching hundreds of thousands. With Digital Green, it’s very localized videos reaching locals within the region.”

Digital Green leverages organizational platforms - N.G.O.’s, companies, government programs â€" that serve rural communities. Partners cover the costs for the video production and screenings or share costs with local farmers. Gandhi says that finding solid partners is Digital Green’s biggest challenge. One such example is PRADAN, which promotes livelihoods for poor families in thousands of villages. Digital Green is also working with the Indian government’s $1 billion National Rural Livelihood Mission. It trains teams in these organizations to produce and facilitate videos - showing them how to identify topics of interest, incorporate scientifically valid information, do basic story boarding, conduct interviews, film and edit videos, and turn it all into engaging discussions.

Now, every fortnight or so, in its villages, after the sun goes down, groups of farmers gather for viewings on house walls or cloth screens hung across verandas. A video may feature a farmer, assisted by an expert, demonstrating the benefits of a technique, or trying something for the first time, or discussing a mistake. The videos cover a wide range of topics: management of seeds and crops, nutrition, home gardening, composting, breast-feeding and so forth (browse them here or here). There is no lugging of television sets and 12-volt car batteries. All that’s required are a battery powered “pico” projector and mini speakers, which fit in a backpack. Transport can be via bicycle, a major logistical advantage.

The viewing process is carefully structured. A local is trained to serve as a facilitator. Typically, the group watches once through, then a second time, with the facilitator stopping and starting, reiterating concepts, soliciting questions, asking people to share experiences, announcing follow-up discussions. Gandhi found that when sessions were actively facilitated, people remained and participated. If not, farmers left quickly.

Digital Green has found that farmers are more likely to adopt new practices if they hear about them from someone of similar socio-economic background, who speaks in the same dialect and accent, and who doesn’t have too much formal expertise. “Farmers will look at whether someone is wearing a watch,” said Gandhi. In one screening, viewers’ interest waned because the farmer in the video had a plastic bucket, a sign of greater prosperity. (This insight could have implications for digital learning more broadly. Should we produce more videos of students teaching peers within their own reference group)

After a video is screened, Digital Green identifies steps which count as adoption. Did you buy or plant the seeds Did you cultivate azolla and use it as animal feed Partners are responsible for gathering this information and reporting back to Digital Green, which painstakingly analyzes data about how many videos were made and viewed, what questions were asked, which practices were adopted, and how farmers fared. Its analytics platform and online mapping tool, Farmerbook, are used to identify things like topics in demand, videos that work well, farmers who are influential, and practices that are spreading.

“Videos are not created equal,” says Gandhi. “We may have 30 videos on how to treat your paddy seeds to increase germination. For some you’ll see a lot of adoption, for some a lot of questions, but no adoption. We use that feedback to inform production.”

There are many things that the system needs to work: inexpensive video technology and partners who produce quality films, facilitate effectively and collect good data. But the real step forward may be applying the simple insight that learning happens best through relationships. Historically, extension has been seen as “a linear transfer of knowledge from scientists and extension agents to the farmers,” commented Ficarelli. “Seventy percent of human communication is misunderstood. This direct contact from farmer to farmer can reduce this degradation.”

Or it may be that people need relevant examples to believe that something is within their grasp. The hardest thing in education is to spark the motivation of the learner. “Human beings are very social,” notes Toyama. “We are influenced by the people around us more than teaching aids.”

Anirban Ghose, director of operations for PRADAN, commented: “When a farmer in India sees a video from the United States, the immediate reaction is: ‘That is something that happens in the United States; it doesn’t happen here.’ But when you show things happening in your community, it challenges the frontiers of what you dare to dream. Adoption is all about psychology. These videos are designed in a manner that they challenge the inner blocks, the psychological barriers people have to adoption.”

Looking ahead, Digital Green is planning to build a Khan Academy-like site for the videos, showing the step-by-step connections across their (now) 2,600 videos. It is planning a pilot with Vodaphone to use audio snippets to reinforce messages. If a farmer forgets something between viewings, he can receive robocalls, call in with questions, or listen to an audio version of a video.

The need for better extension is urgent. Across India, agriculture is changing rapidly. “The monsoon is being progressively delayed each year, so people need to adjust,” notes Gandhi. “Farmers in southern India are seeing diminishing returns from agricultural investments. And people in states like Bihar are switching from wheat to vegetables because of depressed prices and wage inflation.” Everywhere, people are being forced to try new things. “There is a lot of space for improvement,” says Gandhi. As Digital Green continues to gather data about farmers and their agricultural experiments around the country, it will be in a position to get many more farmers asking themselves two questions at the heart of change: Why did those people do that And why aren’t we doing it, too

Join Fixes on Facebook and follow updates on twitter.com/nytimesfixes.

David Bornstein is the author of “How to Change the World,” which has been published in 20 languages, and “The Price of a Dream: The Story of the Grameen Bank,” and is co-author of “Social Entrepreneurship: What Everyone Needs to Know.” He is a co-founder of the Solutions Journalism Network, which supports rigorous reporting about responses to social problems.