Multivariate discriminant analysis: participation in the new extension package programme and its effectiveness on crop productivity in Oromia Region
2005
Samia Zekaria | Habekristos Beyene | Phillip Hollington
Agricultural extension assists farm people through educational procedures in improving farming methods and techniques, increasing production efficiency and income generating to improve their levels of living. This article examines the effectiveness of an extension package programme on the productivity of selected cereal crops in Oromia Region using the 2001/02 Ethiopian Agricultural Sample Enumeration data. The empirical results indicate that the yields of maize and wheat from plots under the New Extension Package Programme (NEP) in the study area were found to be as high as 50% for maize and 39% for wheat compared to yields of the same crops obtained from non-NEP plots. However, the difference was not significant for teff and sorghum. An attempt was also made to examine the relative importance of socio-economic factors that differentiate participant from non-participant farmers of the NEP programme. The univariate statistical test revealed that participants and non-participants differ in all 11 explanatory variables hypothesized to influence the adoption of the NEP programme in the study area. The linear discriminant function analysis showed that programme participants were differentiated using a linear combination of area under improved seed, area under cash crops and quantity of fertilizer. Participants have a larger area under improved seed, a larger area under cash crops, which implies they have higher cash income to spend for the purchase of inputs to increase farm productivity than the non-participants with smaller area under cash crops. Discriminant analysis could then be used to determine which variables are the best predictors of farmers' subsequent adoption choice and helps the implementers to easily classify/identify those farmers who have the potential to adopt NEP.
显示更多 [+] 显示较少 [-]