Video-based Mobile Courseware for Agriculture Extension Workers

Agriculture extension workers engaged by the Government of India to share agricultural best practices among smallholder farmers play a critical role in building the capacity of farmers. To boost their reach and effectiveness, Digital Green trains them on using the community-based, video-enabled extension approach, which involves localized video creation, facilitated screenings and data tracking. To scale this approach in a cost-effective manner as well as with quality and further institutionalize it within the government’s extension structure, we created a mobile courseware, with inbuilt evaluation and accreditation. The video-based curriculum on video production and dissemination as well as agriculture and nutrition topics. The android application has in-built quizzes that test the frontline workers’ knowledge before and after they have completed watching the videos in a course. The scores are recorded and keep track of the individual frontline workers’ proficiency. A certificate issued after completing a course (on gaining a minimum of 70% marks) incentivizes the frontline workers.

With a grant from Oracle, Digital Green is currently expanding access to its mobile courseware on improved agricultural practices. This grant is supporting the development of the mobile curricula to train 800 extension workers who will further train 80,000 farmers on practices that increase agricultural productivity. This work is being carried out in Jharkhand, Andhra Pradesh and Odisha. 

Jharkhand

Frontline workers in Jharkhand watching videos on their smartphones and learning how to use the platform.

In Jharkhand, our team had rolled out courses in 22 blocks spread across 12 districts of Jharkhand in 2018 among 330 frontline workers (FLWs) (known as Aajeevika Krishak Mitras-AKMs/Community Resource Persons-CRPs in Jharkhand). These courses were on potato and pigeon pea cultivation. Based on a comparison of FLW knowledge levels before and after the course, and anecdotal feedback from the field, we found this training app to be user-friendly and effective.

Based on the previous years’ experience, in 2019, our team rolled out 14 new courses on potato and pigeon pea cultivation, non-pesticidal management (NPM) and agri-nutrition practices across the same geography. In this cycle, the team has integrated knowledge assessment within the mobile app replacing the pen and paper-based tests used in the previous year. This would enable the team to effectively assess knowledge transfer to FLWs through this intervention. The mobile courseware is built on a Community Training platform developed by Microsoft Research and our team worked with Microsoft Research to improve the application’s architecture so that the courses can be accessed through different Android phones and videos can be streamed even in limited connectivity zones.

Odisha

We reached out to farmers across Keonjhar district in Odisha to understand the demand for self-service videos. Increase in smartphone penetration, fibre connectivity, ease of understanding through a video over conventional means and need for timely advisories were factors in favour of the experiment. 

Lalita Mohanta of Village Bhalughara, Odisha, a farmer, watching short videos shared on the WhatsApp group at her home.

Based on the need and enthusiasm amongst the farmers for such a solution, the team is designing small video clips on paddy cultivation practices such as blast (fungal infestation) management, harvesting and drying. Each video focuses on one micro practice and is available on Digital Green’s YouTube channel. Our team collected the phone numbers of farmers who own a smartphone and added them on WhatsApp groups managed by the block-level officer of the Department of Agriculture. Links to the YouTube video clips are shared on the WhatsApp group at the precise time a farmer would need to adopt a particular practice.

There are nearly 900 farmers across 10 Gram Panchayats under the block level officer and our team is tracking viewership of the video clips on YouTube to assess the demand for self-service content by farmers owning smartphones. They are also organizing focused group discussions with the farmers to see if knowledge is being shared across the peers and any of these being translated into the uptake of practices.

Uncovering the Link Between Access to Nutrition and Markets Through Research

According to the National Sample Survey Organisation (NSSO), average fruit and vegetables (F&V) consumption in Bihar equalled 132 g/capita/day in 2011-2012. Therefore, people in Bihar on average consume approximately one-third of the global recommendation of 400 grams/capita/day (FAO and WHO, 2014). In turn, consumers dependent upon nutritionally vulnerable markets (i.e. those that are traditional, small and often rural) are likely to face the greatest challenges to fruit and vegetable access and affordability.

Digital Green’s ‘Loop’ project aims to help farmers save time and reduce the cost of transporting their vegetable produce to market. In Loop, an aggregator collects and markets fruits and vegetables on behalf of farmers – optimising transport based on the volume of produce. The model leverages digital technology to ensure transparency, efficiency and trust between the aggregator and farmer, which in turn enables better negotiation power for bulk selling in most cases. It also saves the farmer market transport costs and allows them to reinvest the time and money into on-farm and recreational activities.

The Market Intervention for Nutritional Improvement (MINI) project, funded by the Bill & Melinda Gates Foundation (BMGF) and the UK Department for International Development (DfID), aims to explore the nutritional aspects of the market and technical interventions in vegetable and poultry value chains in Bihar and Bangladesh. Led by the School of Oriental and African Studies (SOAS, University of London) and the International Livestock Research Institute (ILRI, Vietnam), the project involves researchers from the London School of Hygiene and Tropical Medicine (LSHTM), Bangladesh Agricultural University (BAU, Dhaka) and Lincoln University, New Zealand, as well as Sadman Sadek, Vinay Rana, Dr Nadagouda and Namita Singh from Digital Green (amongst others!).

Digital Green’s Vinay Rana (Patna office) works with farmers to map the locations of Loop villages and markets during the first ‘Group Model Building’ session in Muzaffarpur (January 2019)

The research focuses on the ways in which aggregation schemes (such as Loop) may be made more nutritionally sensitive – which refers to the availability and affordability of F&V in some of the more rural, traditional and retail-based markets. However, achieving this goal is not a case of simply sending greater volumes of F&V towards smaller markets. This is owing to various conditions and feedbacks that make rural markets less attractive to farmers and consumers alike, including less developed transport and market infrastructures, inferior market capacities and the typically weaker purchasing power of rural consumers. As a consequence, we need to think carefully about how the aggregation system such as Loop may be utilised to minimise trade-offs and achieve ‘win-win’ scenarios for both farmers and consumers.

The MINI project takes a multidisciplinary approach to the research problem. To date, rapid value chain analyses in Bihar, India and Jessore district, Bangladesh, have identified the key actors involved in the respective horticultural value chains. From here, survey campaigns interviewed 360 farming households in both Bihar and Bangladesh. In Bhojpur and Muzaffarpur districts, Bihar, we tentatively find that Loop farmers on average produced and sold higher quantities of F&V relative to farmers that had not participated in Loop between 2018-2019 (please note: this result is not final and yet to be tested with robust econometrics). Through common econometric techniques, the research team is currently investigating the extent to which these outcomes can be attributed to Loop participation, as well as the various socioeconomic factors that might determine Loop participation in the first place.

A team member from the Centre for Media Studies (CMS, Delhi) pilots the household survey with a farmer from Minapur block, Muzaffarpur (February 2019)

System dynamics modelling builds upon the value chain analyses and household surveys. Working alongside the experts at Digital Green and select groups of farmers, traders and commission agents in Bhojpur and Muzaffarpur, the MINI project team have built a system dynamics model to simulate the future evolution of Loop under various scenarios. These scenarios range from internal changes to Loop (e.g. scaling-up farmer numbers and transport subsidies) to making changes within the wider enabling environment – such as the introduction of accessible cold storage facilities in F&V markets. The model is just starting to produce the first future simulations as we speak, so please watch this space over the coming weeks and months!

The start of 2020 sees the MINI project beginning to apply the same techniques to the homestead poultry programme run by JEEViKA in Bihar. Whilst JEEViKA’s poultry programme initially focused on improving household nutrition (e.g. through chicken meat and egg consumption), the MINI project and Digital Green are interested in the potential market linkages and opportunities to expand the poultry programme for nutritional benefits beyond the household scale. This analysis will run in parallel with a similar project based in Ghana, led by Dr Karl Rich (ILRI), to ultimately evaluate whether a pioneering aggregation model such as Loop might bring about nutritional benefits in poultry value chains and markets.

 

About the Author:

Dr Gregory Cooper is a Postdoctoral Research Fellow at the Centre for Development, Environment and Policy (CeDEP), SOAS, University of London. Follow the story of the research project here on his blog.

A more detailed interview on the MINI research project can be found here in a series of 3 articles: Unravelling value chains through participatory modelling: Part 1, Part 2 & Part 3