Improving smallholder agriculture via video-based group extension.

Area of Science:

• Agricultural Extension and Technology Adoption
• Climate-Smart Agriculture
• Development Economics

Background:

• Scaling agricultural advice presents significant operational challenges.
• Technology integration in agricultural extension risks reduced efficacy due to limited customization and human interaction.
• Digital content delivery, such as videos, may improve learning and adoption if content is reinforced.

Purpose of the Study:

• To evaluate the efficacy of video supplements to standard human-provided extension services for promoting System Rice Intensification (SRI).
• To assess the impact of digital interventions on adoption rates and agricultural output among female farmers in India.
• To investigate the influence of specific content reinforcements (labor costs, role models) on treatment effect estimation.

Main Methods:

• A randomized controlled trial was conducted, supplementing standard extension services with video content for female farmers in India.
• Weighted quantile regressions were employed to address non-normally distributed outcomes and analyze effects across different yield quantiles.
• A Bayesian hierarchical model was utilized for robust estimation of treatment effects, with sensitivity analysis to various priors.

Main Results:

• Overall treatment effects were large but imprecise due to right-skewed outcome distributions, particularly among high-yield farmers.
• Quantile regressions (25th and 50th percentiles) and the Bayesian model indicated significant positive treatment effects on output.
• Subtreatments reinforcing labor cost information and presenting role models demonstrated even higher estimated treatment effects on output.

Conclusions:

• Video supplements can be an effective tool to enhance agricultural extension services, particularly for promoting climate-smart practices like SRI.
• Addressing specific farmer concerns (e.g., labor costs) and leveraging social influence (role models) through digital content can amplify positive impacts.
• Careful consideration of outcome distribution and analytical methods is crucial for precise estimation of treatment effects in agricultural interventions.