Factors Influencing the Adoption of Improved Rice Varieties in Rwanda: An Application of the Conditional Logit Model (CLM)

This study investigates the factors influencing the adoption of improved rice varieties by small- scale rice farmers in Rwanda. The Conditional Logit Model (CLM) was applied to explore the variety attributes and farm household characteristics influencing the farmers’ choice among different alternatives. The model is based on random utility maximization and makes the probabilities of the choice of variety dependent on variety-specific attributes and farm household-specific characteristics. A sample size of 180 rice farmers interviewed was randomly selected from six rice growing marshlands. Results revealed that a majority of rice farmers (78 %) had adopted improved varieties but the proportion of land cultivated to these varieties was still low (ranging from 0.05 to 0.2 ha). The maximum likelihood analysis showed that the prices of seeds and of paddy and the yield are variety-specific attributes that significantly influence the farmers’ choice of an improved rice variety. The higher the price of seed, the lower the likelihood of adoption; the higher the price of paddy and yield, the higher the likelihood of adoption. Also, farm size, labor availability within a farm household, and access to financial facilities significantly influence farmers’ decisions to adopt improved rice varieties. The findings suggest that future policies should make efforts to improve the seed supply system among farmers to increase the intensity of use of improved varieties. In addition, the prices of paddy should provide a market incentive that increases the likelihood of adoption through market development.